About Search Language and Strategies

You may be used to searching in Google, but the wording you use to search a library database will need to be somewhat different. Here are the differences in search language between Google and a library database (or discovery layer):

• In Google: You can type a question in the same way you might ask it in a conversation, for example: What are the health benefits of regular exercise? This is called searching using natural language.
• In a library database:
  • Keywords: You should search using keywords, which are the words that hold the key idea of what you are trying to find. Learn more about Keywords.
  • Boolean Operators: You can combine or exclude your keywords in your search using words called Boolean Operators. Learn more about Boolean Operators.
  • Wildcards: You can search for words with spelling variations by replacing a letter with a wildcard symbol. Learn more about wildcards.
  • Truncation: You can search for variations of a word by adding a truncation symbol after the root of the word. Learn more about truncation.
  • Subject Headings: You can also use phrases called “subject headings,” which are the standard terms for specific topics, to help you find what you’re looking for. Learn more about subject headings.

Explore the tabs to learn more about the search language used in library databases.

How Do I Use Keywords?

The words you choose for your search will have an impact on the results you find. There are many different ways of describing the same topic. Doing the same search multiple times with a variety of terms will yield the most complete results.

Take a moment to learn more about Keywords, Synonyms and Related Terms.

Keywords

Keywords are words that hold the essence, or the key idea, of what you are trying to find. Keywords are usually nouns, e.g. people, places or things. Don’t include words like why, what, where, when, if, the, etc. in your database search. Using relevant keywords in your search will lead you to better information. You can identify the first keywords from the topic itself. Try to think of 2 - 4 keywords. If you have too few, your search results won’t be specific enough, and if you have too many, you may get too few results.

Example:

• Topic: Are text message reminders effective in maintaining lower blood sugars among teenagers with Type 1 diabetes?
• Keywords: text message reminders, blood sugars, teenagers, Type 1 diabetes

Synonyms

Synonyms are words that have the same, or similar meaning as the main keywords. Synonyms of keywords are interchangeable, which means that the meaning of your search will remain the same. Synonyms can be used to broaden your search to retrieve more results.

Here are a few examples of synonyms for the topic: Are text reminders effective in maintaining lower blood sugars among teenagers with Type 1 diabetes?

• Text message reminders: SMS reminders
• Blood sugars: Blood glucose, glycaemia
• teenagers: youth, high school-aged
• Type 1 diabetes: juvenile diabetes

Related Terms

Related terms are words that generally mean the same thing as the main keywords. For example: tablet - device, car - vehicle, pop - carbonated drink. While related terms don’t mean exactly the same thing as the words they are replacing (synonyms), they are an excellent tool for broadening the scope of your search.

Here are a few related terms for the topic: Are text reminders effective in maintaining lower blood sugars among teenagers with Type 1 diabetes?

• Text message reminders: email reminders
• Blood sugars: hypoglycemia, hyperglycemia
• teenagers: children, young adults
• Type 1 diabetes: autoimmune disease

What Are Boolean Operators?

Boolean Operators, including AND, OR, and NOT, are words that make it easy for you to customize the results of your search. Watch the video and read the information below to learn more about Boolean Operators and how to use them.

When searching for information in a library database or a search engine, you may want to combine some keywords, or exclude certain words, to ensure that your search results are more focused or relevant to your topic.

TIP: In library databases, you don’t need to capitalize proper nouns (e.g. Twitter, Trudeau or Ontario), but Operators must be typed in all capital letters, e.g. NOT, AND, OR.

AND

Using the AND operator tells the database that all words, or terms, that you have connected with AND must be found in any results returned.

If, for example, you are searching for articles about marketing with Twitter, you could search for: marketing AND Twitter. Only articles that include both Twitter and marketing will be in your search results.

OR

For a broader search, to find articles that discuss marketing with Facebook or Twitter, you could use the Operator OR, e.g. marketing AND (Facebook OR Twitter). The results from this search will include articles that talk about marketing and Facebook, or marketing and Twitter, or marketing and Facebook and Twitter.

In the above example, you'll see that brackets are included in the search. When you are using more than one Operator in a search (e.g. AND and OR), you will need to group your keywords and operator words using brackets, so that the database knows which action to perform first.

NOT

The Operator NOT will narrow your search results by excluding or removing a specific word or words from the search results.

For example, if you’re researching marketing but are not interested in articles about marketing using Twitter, you could search for: marketing NOT Twitter.

Your results from this search will not include any articles that contain the word Twitter.

Be careful when using NOT, as it can remove results that would actually have been relevant. For example, if you were searching for articles about adolescents, adding NOT children to your search would remove all results that address both adolescents and children - and even articles with titles such as “Treatment X improves sleep in adolescents but not in children.”

How Do I Search for a Phrase?

Sometimes you might want to search for a phrase, where the words always appear together and in a specific order. Watch the video and read the information below to learn more about how to search for phrases.

When searching for an exact phrase, (i.e. exactly the same words in the same exact order), most library databases support the use of "quotation marks" (“ “) around the phrase, which could be two or more words.

Quotation marks instruct the database to return results that include that exact phrase. Searching for an exact phrase can help to reduce the number of irrelevant results.

For example, if you search for articles about body language (without quotation marks), your results will include both words, but body might be on the first page of the article, and language on the last.

Searching for “body language” will only return results that include that exact phrase. Using quotation marks to search for an exact phrase will narrow down your results and make them more relevant. Exact phrase searching with quotation marks will also work in Google.

There are, however, a few databases that do not recognize or support exact phrase searching using quotation marks. If you get no results, or too few, remove the quotation marks from around your phrase and search again.

What Are Wildcards?

A wildcard is a special character that replaces one or more letters in a word (e.g. colo#r) in order to search for multiple variations of the word. When keyword searching, you may miss relevant and useful results if the term you have searched for does not appear in that exact form in an article or book. Wildcard symbols can help you to find word variations so that you don’t miss anything.

To use a wildcard, insert the wildcard symbol used by that database to replace the letter that may change.

For example, if you do a search for pediatric and an article uses paediatric instead, that article won't be in your search results. If you searched using a wildcard, you could search for p#diatric and see results for both pediatric and paediatric.

Databases use different symbols for their wildcard. Common wildcard symbols used in different databases include:

• the question mark (?)
• the pound sign (#)
• the dollar sign ($)
• the percentage symbol (%)
• the exclamation mark (!)

If you want or need to use a wildcard in your search, check the help section in the database you’re using to find the wildcard options.

What is Truncation?

To ‘truncate’ a word simply means to shorten it by removing one or more letters to go back to the root word. Truncating a word allows you to search for multiple variations of a word at once. You can do this by adding a truncation symbol (e.g. *) to the end of the root of the words. Like with wildcards, different databases use different symbols for truncation; check the database help to find out which one to use. Watch the video and read the information below to learn more about how to use truncation.

Let’s imagine we are searching for articles about the practice of nursing in Canada. If we search for those keywords only, nursing AND Canada, we may miss articles that include the words nurse, nurses, Canada’s or Canadians. If we truncate each word, we will get more results. So, to search nursing in Canada, we could truncate both keywords and search for: nurs* AND Canad*.

Tips for truncating:

• Be careful not to truncate the word too much.
  Truncating too many letters from a word can lead to results that are unrelated to your intended search term. For example, searching for nursing in Canada using the truncated word nurs* will also bring back results about nurseries, and truncating Canada to Can* may retrieve articles with words like cantaloupe, cancer, canned, etc.
• Google Scholar uses “automatic stemming”.
  This means it automatically treats your search terms as if they have a truncation symbol at the end, so a search for nurse would also return nurses, but not nursing.
• Some databases have special rules around truncation.
  For example, you may or may not be able to use truncation inside a phrase (with quotation marks), and there may be a limit to the number of characters you must type before the truncation symbol (in PubMed, you must type at least 4 characters, so you would not be able to truncate nur*).

Database Help Pages

Databases typically have help pages and/or user guides to help you use that database. This is also where you can find information on wildcard and truncation symbols. See the information on the help pages for some common health sciences databases below.

In PubMed, you can find the help page by clicking FAQ & User Guide, which will be located either under the search bar, in the section labelled “Learn.”

In CINAHL, you can find the help page by clicking Help, located in the top right of the page.

In Google Scholar, you can find the help page by clicking Help, located in the bottom right of the page.

What Are Subject Headings?

Many databases provide the option of searching using subject headings. A subject heading is a word or phrase that is assigned to an article or other resource and describes the topic of the resource. It is a bit like a social media hashtag, but without the # symbol.

Subject headings use a controlled vocabulary, meaning that there is a list of possible headings that everyone who assigns them must agree upon. A single article might have multiple subject headings, especially if it covers a range of topics. Watch the video or read the information below to learn more about subject headings.

Why Use Subject Headings

Searching by subject heading is more powerful than searching by natural language or even by keyword. When you search by keyword, that word might show up in a number of places in relation to the article, such as the abstract, the author’s name or the journal title, even if it’s not specifically what the article is about. This means your search results may include many articles that are not relevant to you. Conversely, when you search by subject heading, you are limiting your results to articles that are actually about that topic.

Not every concept has a subject heading, so in some cases you won’t be able to find a subject heading that means exactly what you’re looking for. Rather than trying to convert every keyword into a subject heading, it’s often best to use a combination of keywords and subject headings.

Subject headings are also useful because they are arranged in a hierarchy of broader and narrower terms, sometimes called a tree structure. Exploring the tree may give you ideas for search terms you hadn’t considered.

Example

If you search Nurse-Patient Ratio as a subject heading and don’t find many results, you could look at the tree structure, find the broader term Personnel Staffing and Scheduling, and try that one instead. Here is what a portion of this tree looks like in CINAHL:

• Health Services Administration
  • Management
  • Personnel Management
  • Personnel Staffing and Scheduling
    • Flexible Scheduling
    • Nurse-Patient Ratio
    • Personnel Staffing and Scheduling Information Systems
    • Shiftwork
    • Skill Mix
    • Understaffing
    • Work Assignments

Subject Heading Systems (MeSH, etc.)

Different databases use different subject heading systems, which means you will need to look up your subject headings in each database you use. Here are two common systems:

• Medical Subject Headings (MeSH): This is one of the most common systems. It is used by PubMed, Medline, Cochrane Library, and others.
• CINAHL Subject Headings: This is the system of subject headings used by CINAHL.

How to Search with CINAHL

Watch the video or read the instructions below to find out how to conduct a search in the CINAHL database. Visit your library website to access CINAHL.

To search in CINAHL:

1. Once you have opened CINAHL, click Advanced Search.
2. Make sure “Suggest Subject Terms” is checked, type in your search term (e.g. privacy) and click Search.
3. Click on a relevant subject heading in the list, e.g. “Privacy and Confidentiality.”
4. The tree view will appear; this view shows the broader and more specific terms for the subject. Select an additional term if it would be more appropriate for your search.
5. Click the icon under the Scope column and read the Scope Note to make sure the subject heading means exactly what you think it means.
6. Click the checkbox beside the term you want to use, e.g. “Privacy and Confidentiality”
7. To the right of the Scope column, you’ll see a list of subheadings. If you see one or more that are particularly relevant to your search, you can select them now. Note: This will limit your search results. We suggest starting by leaving the “Include All Subheadings” box checked.
8. Below the Search Database button, you can select either of the following two options:
   1. Explode: This option will give you more search results by including all narrower terms below your subject heading in the tree (in this example, Duty to Warn and Genetic Privacy).
   2. Major Concept: This option will give you fewer (but potentially more relevant) search results by limiting your search to articles where your subject heading has been flagged as a major concept, meaning it is especially important to the article.
9. Click Search Database. You will see a list of results.

How to Conduct a Search using PubMed

Watch the video or read the instructions below to find out how to conduct a search in the PubMed database.

To search in PubMed:

1. Open PubMed and click MeSH Database.
2. Type your search terms into the search box and click Search. Your search results will show a list of MeSH headings related to your search term. Each comes with a brief definition to help you know which term you want to use.
3. Select a MeSH heading from the list. This could be the same term that you originally searched, or you could see one that is a better option. This will display the MeSH record for that heading.
4. Modify your search using any of the following options included in the MeSH record:
   1. Subheadings: Select any subheadings that would be relevant to your search. Be aware that this would limit your search results.
   2. Restrict to MeSH Major Topic: Check this box to see fewer (but potentially more relevant) search results by limiting your search to articles where your MeSH heading has been flagged as a major topic.
   3. Subject headings in the “tree view”: You can see broader and narrower terms related to your MeSH heading and select one of them if they would be more appropriate for your search.
5. To run your search, in the “PubMed Search Builder” in the top right of the page, click Add to search builder. Your search term will appear in the box.
6. Click Search PubMed to run the search. Your search results will show all the articles that have the term you chose as a major MeSH heading. Note: by default, in addition to the MeSH heading you’ve selected, PubMed also searches more specific terms beneath that heading in the tree structure.
7. Refine your search as needed by adding additional terms, either as subject headings or as keywords. You can also try adding filters to your search; learn more about how to use filters in the module Filter Your Search Results.

About Filters

Discovery layers and individual databases have filters (sometimes called “limits” or “limiters”) that help you refine your search so that you retrieve results that are more accurate and relevant to what you are looking for. When you select a filter, all search results that do not match the filter’s criteria will be removed from your search results. Using filters can help you narrow down your search results to the ones that are most relevant to your topic. Explore the tabs to find out about the following types of filters:

• General filters (standard filters in most databases and discovery layers)
• CINAHL filters
• PubMed filters

General Filters

Some filters are used in almost all databases and discovery layers. Knowing about these filters can help you narrow your search in many different search interfaces. Watch the video and read the information below.

Here are some commonly used filters:

• Available Online/In Your Library/Full Text. This filter ensures that you find sources that you can view in their entirety (the full text of the article, etc.). There are various ways of labelling this filter, so look for any of these three.
• Peer Reviewed. This filter ensures that all results are from peer-reviewed journals.
• Date Range/Publication Date. This filter limits your search to material published within a particular time frame.
• Language. This filter limits your search results to materials written in one or more languages. You can select all languages that you can read. In CINAHL, hold the control key (command key on a Mac) to select multiple languages.
• Age. This filter limits your search to articles about specific age-based populations (infants, children, adolescents, etc.). This filter is most commonly found in health science-focused databases; it may not appear in more general databases.

Note: Databases do not typically have filters for geographic locations (such as Canada or Ontario). To search for articles that refer to specific locations, you will need to use keywords and/or subject headings. See the module Conduct Your Search for information.

Using Filters in CINAHL

When searching CINAHL, you need to select your filters before performing your search. Start by clicking on Advanced Search (under the search box) to display all of the search options. Watch the video and read the information below to learn about the filters you can use in CINAHL.

Filters

In addition to the general filters, you can also use the following filters in CINAHL:

• First Author is Nurse. Use this filter to search for articles written by nurses. If using this filter returns too few results, try switching to Any Author is Nurse.
• Any Author is Nurse. Use this filter to search for articles where a nurse was part of the research team.
• Evidence-Based Practice. Use this filter to limit your search results to Evidence-Based Practice (EBP) research articles, including the following types of sources:
  • Meta-analyses
  • Systematic reviews
  • Articles about EBP
  • Commentaries on research studies
• Publication Type. Use this filter to limit your search results to particular types of studies. Hold down the control (or command) key to select multiple options. Options include the following types of publication:
  • Systematic reviews
  • Randomized controlled trials
  • Commentary

Tip: There is no filter in CINAHL to search for either qualitative or quantitative studies. To limit your search to qualitative or quantitative studies, add the CINAHL subject heading Qualitative Studies or Quantitative Studies to your search.

Using Filters in PubMed

When searching in PubMed, select your filters on the search results page after performing your search. Watch the video and read the information below to find out how to use filters in PubMed.

How to use filters in PubMed

1. Perform a search. You will now see the basic filters on the left side of the page.
2. Click Additional Filters to see more options.
3. Select any filters that you want to add, and click Show. This will add those additional filters to the list on the left of the page. Note: this step will only add these filters as options. If you want to use them in your search, you must select them again in the next step.
4. Select all of the filters you want to use to narrow your search from the list of filters on the left.

Filters

In addition to the general filters, you can also use the following filters in PubMed:

• Article Type. Use this filter to limit your search results to particular types of articles. Check as many boxes as you like. Some options are available on the results page by default, but you must click Additional Filters to see them all. Options include the following types of articles:
  • Systematic reviews
  • Randomized controlled trials
  • Meta-analysis
• Publication Date. There are two options for filtering by publication date.
  • Results by Year: This filter provides a slider interface to select a range of years.
  • Publication Date: This filter allows you to search for articles published in the past 1 Year, 5 Years, 10 Years, or a Custom Range.
• Full Text. Use this filter to limit your search results to articles for which you have access to the full text. This filter includes both articles from journals your library subscribes to and articles that anyone can access for free.
• Free Full Text. Use this filter to limit your search results to articles for which anyone can access the full text for free. Use this filter if you are not currently studying or working at an institution with journal subscriptions. 

Tip: There is no filter in PubMed to search for either qualitative or quantitative studies; however, the following methods will allow you to limit your results:

• Qualitative studies: Add the MeSH heading
• Qualitative Research to your search. Quantitative studies: Choose specific study types from the Article Type filter.

Note: Almost all of the journals in PubMed are peer-reviewed, so there is no filter for peer review. If you absolutely must confirm whether a journal is peer-reviewed, go to the journal’s website and look for an About page. If it’s peer-reviewed, it should say so on this page.

When you first click on the title of a result in your search, you won’t see the actual text of the article. Instead, you’ll see a page called the record, which will have information about the author, the journal title, the abstract, and more. From the record, you can do the following:

1. Read the record for information about the resource. The record should give you enough information to decide whether the resource will be useful to you.
2. Click a link to see the full article (this is what we call the full text).

Explore the tabs for information about the following aspects of accessing the full text of articles:

• Opening the full text from the record
• Choosing the format for viewing the full text, if you have more than one option (HTML or PDF)
• Requesting an interlibrary loan to access resources that aren’t available through your library

How to Open Full Text

Once you have opened a record, you will need to click on a link to open the full text. Records in different search interfaces will have links in different places. Watch the video and read the information below to find out where to find the full text link in various search interfaces.

CINAHL

To access full text in CINAHL, click the full text link, which is typically in the upper left corner of the record.

PubMed

PubMed only contains records, so to access the full text, you will need to go from PubMed to a page containing the article. To access the full text through PubMed, follow these steps:

• Click one of the full text links in PubMed, which are typically in the upper right corner of the record. These links will not take you straight to the full text. They will take you to another page (sometimes in PubMed Central, which is a section of PubMed that contains open access articles).
• Look for another full text link and click to open the full text.

Google Scholar

You don’t have to open the record in Google Scholar to see the full text link; you can access them from the search results. Google Scholar provides full text links for the following:

• Full text that it is able to find from a public website.

You will find these to the right of the search results.

• Full text from your library if you set it up to do so.

​​​​​​​Google Scholar may be able to provide full text links for articles that your library subscribes to if it knows you are a student (e.g. sometimes if you are on campus); however, by default, Google Scholar does not include these links. You can change your settings to view your library links as long as your library has enabled this function.

To set up library links in Google Scholar:

• Click the hamburger menu icon in the top left (see Figure 4).
• Click Settings.
• Click Library Links from the menu on the left.
• Type in the name of your school and press enter.
• Check the box next to your school’s name and click Save.
• After completing these steps, you will see your library’s full text links for articles your library subscribes to.

​​​​​​​​​​​​​​Note: If you have trouble with this process, you can also copy and paste the title of an article you find in Google Scholar into your library’s discovery layer or databases to see if your library provides access to that article.

Choosing a full text format

The full text of an article may be available in different formats: HTML, PDF, or both. The content of the article is exactly the same between the PDF and HTML versions. Here is some information about each format.

• PDF: This is the most common format for full text articles. It’s great for saving the article to your computer or for printing it. PDF articles may appear in your browser, or you can open them in software such as Adobe Acrobat Reader.
• HTML: HTML is the same format used on most websites. An HTML article will appear in your browser. In some databases, the HTML full text will appear on the page below the record, while other databases will have a link to the HTML full text.

How do I choose?

• If you want to save or print the article, choose PDF.
• If you want to quote the source, choose PDF to get a page number for your citations.
• If you want the text to flow and resize to fit the size of your screen (especially on mobile devices), choose HTML.

Interlibrary Loan (ILL)

Regardless of which search interface you’re using, you may find a record for an article that interests you, but you can’t access the full text. If this happens, contact your library and ask whether they can help you get a copy of the full text.

Accessing a resource from another library is known as an interlibrary loan (ILL); your library may not use that name, but most college libraries offer an ILL service. The library staff can request a copy of the full text from another library, which they can then provide to you. The article may arrive in just a couple of days, but it could take a week or more, so be sure to start your searching long before your assignment is due. Typically, the library would send you the article as a PDF.

TIP: Be sure to use your student email account (and not a personal email account) when contacting your library, so they know right away that you’re a student.

Introduction to Documenting Your Search

Documenting your search usually means writing down all the steps you went through during your search process, including information such as the following:

• which search terms you used
• which databases you used
• how many results you retrieved

Why document your search?

Whether or not it’s required for your assignment, documenting your search is useful because it helps you keep track of which databases and which search techniques you’ve already tried. When you know what you’ve already tried, it’s easier to try other strategies to find what you need. You may want try one of the following approaches:

• Trying a new combination of search terms
• Trying the same search terms in a different database
• Adding or removing filters from searches

How do you document your search?

If your professor requires that you document your search, they may or may not provide a format for you to follow. No matter what, read your assignment outline carefully and follow all the instructions.

This module will show you two methods for documenting your search:

• Accessing your search history
• Creating a PRISMA flow diagram

Explore the tabs to learn about each of these methods.

Note: If your instructor has not asked you to document your search but you want a good way to keep track of all the materials you’ve found, try a citation manager. You can learn more about citation managers in the Use a Citation Manager module.

Accessing Your Database Search History

One way to document your search is by accessing your search history in the databases you’ve used. You can save that history in one of the following ways:

• If you are required to document your search for an assignment, you can copy and paste your search history into your assignment.
• If you’re just keeping track on your own, you can save the search history in a document or a folder.

If you’ve searched more than one database, be sure to include your search history for all of them. Watch the video and view the information below to learn how to find your search history in CINAHL and PubMed. Other databases will likely have the search history in similar places; this module will help you understand where to look so it’s easier to find elsewhere.

CINAHL

To access your search history in the CINAHL database:

1. Perform your searches in CINAHL.
2. When you are finished searching, click Search History below the search box.
3. Click Print Search History to view a simplified version.
4. Copy the simplified history and paste it into a document.

PubMed

To access your search history in PubMed:

1. Perform your searches in PubMed.
2. When you are finished searching, click Advanced under the search box.
3. Scroll down to see your search history under History and Search Details.
4. To see a search in PubMed format, click on the chevron (arrow) in the Details column.
5. To download your search history as an Excel document, click Download.
6. Copy and paste the information from the Excel file into your assignment or other document, or save the file in a folder with other search histories.

Creating a PRISMA flow diagram

A PRISMA flow diagram is a specific format for documenting your search. It was designed for researchers performing systematic reviews and meta-analyses, but your instructor may ask you to include one with your assignment.

Note: If your instructor has NOT asked you for a PRISMA flow diagram, you can skip this tab.

How to create a PRISMA flow diagram

To create a Prisma flow diagram, you can use the resources and information below:

• PRISMA flow diagram templates
• PRISMA 2020 Statement
• PRISMA 2020 Explanation and Elaboration

Note: If your instructor has asked you to use a modified version, you should follow their instructions.

Sample

This is what a standard PRISMA flow diagram looks like (Figure 1), using the 2020 version known as PRISMA 2020. You can find templates for the PRISMA flow diagram on the PRISMA website.

Figure 1: PRISMA flow diagram. The image of the PRISMA flow diagram is a derivative of the PRISMA Flow Diagram template from the PRISMA Statement on the PRISMA website is used under a CC BY license.