Search information
Introduction
You have your search questions and search terms listed and you have chosen your sources of information. Now you can really start searching. In this module you will learn a number of techniques to search more effectively:
- Combine & exclude
- Exact phrase
- Trunking
- Masking with wildcards
- Using keywords
- Citation tracking
Finally, you'll get an overview of tips to improve your search results. What can you do if you have too few good results? What can you do if you have too many results?
Look carefully at the different options before you start searching. The search options determine the way you search.
Simple or advanced search
Most databases and search engines have a simple search bar on their home page. In this search bar you can enter a single search term, or you can combine search terms, for example:
school AND amsterdam
Usually there are also more extensive options. These can be found under advanced search. This is also called advanced search or multi-field search.
With advanced search you can formulate your search query much more precisely, combine or exclude terms. You can enter more search terms at once, in different search fields.
Example: Advanced search in CataloguePlus
SEARCH TECHNIQUES
Illustration by Jurriaan Gorter, published under a Creative Commons Attribution 3.0 Netherlands license
When you start searching with a search engine or in a database, you often just type in some words. For simple questions, you will usually find the answer you are looking for that way. However, this is often not enough.
Use the extra features that search engines and databases offer you and you will get a much better result.
Because a search engine or database does not understand "ordinary human language," you have to adapt the search query.
You will learn in this module a number of techniques you can use to do this.
... Combining words
Relationships between search terms
If you are going to search with multiple terms at once, you need to specify exactly what the relationship between the search terms is..There are three basic relationships between search terms:
Both terms must appear in the search result (AND), for example soccer AND Amsterdam
At least one of the terms must occur (OR), for example soccer OR tennis
This is also useful for synonyms, e.g. children OR kids
One term must be excluded (NOT), e.g. reading NOT disorder
Graphically, the relationships between search terms are shown like this:
Sometimes you have to indicate the relationship between search terms yourself by typing AND, OR and NOT between the search terms. In Dutch-language databases you may have to use EN, OF and NIET.
In most databases and search engines you will find a search form in Advanced Search where you can choose one of these options. Often you can also combine the different options.
Example with Web of Science
You are looking for information on the influence of social media on young people. You want to look up publications that contain both terms.
Suddenly you realize that adolescents is an important alternative term for teenagers. You adjust your search so that the term social media appears in combination with at least one of the terms teenagers or adolescents.
Note: If you combine with both AND and OR then you must place the OR combination within one search field in most advanced search forms (above image). If you don't, the search engine won't know which search terms to combine first and you won't always get the results you want.
Right now, you're not interested in articles about marketing. You want to find articles about social media combined with the terms teenagers or adolescents, but without the terms marketing or advertising.
Your search command results in:
(food OR breakfast OR appetite OR eating OR menu OR vegetables) AND (aged OR elderly OR dementia OR "nursing home") AND ("quality of life" OR social OR well-being OR loneliness OR tasty)
Nice to know: Nesting terms
In many databases and some search engines you can also formulate more complex searches: by nesting terms. To indicate what to combine with AND and what to combine with OR and in what order, you can nest terms. You do this using round brackets.
Example:
If you search for
social media AND (teenagers OR adolescents)
you will find documents about teenagers and/or adolescents and social media.
If you remove the parentheses
social media AND teenagers OR adolescents
you not only get documents on social media and teenagers, but also documents on adolescents without social media in them.
Combine sets
In come up with search terms, you have seen that each search query consists of more components. For each component, you created a set of appropriate search terms.
The sets consist of word variants, such as translations and synonyms. You connect the terms within a set with OR, because you want to find at least one of these variants. You connect the sets themselves with AND, because you want to find all of them.
In this example, we combine variants of the terms for elderly with variants of the terms for nutrition and its social aspects. This is a schematic representation of this search query:
Je zoekopdracht wordt dus:
... Exact phrase
Combining words using quotation marks
Sometimes you want to search for an exact word combination.
You can put word combinations or whole sentences between quotation marks. This combination or sentence must then appear exactly as it appears in the document you are looking for. This is very useful if you are searching for a name or a compound term.
Caution! You may miss articles that are about your topic. For example, you search for the exact word combination
"online lecture"
With this combined, you won't find articles with
"lecture in an online learning environment"
You do find those articles with the search
online AND lecture
You can also specify that search terms must occur a certain distance apart. That can make a huge difference in the noise between your results.
Nice to know: proximity operators
To make sure you only find documents where your search terms are close together, you can use so-called proximity or adjacency operators. How this works varies by database. Below are the most common ones:
NEAR/x
| example |
online NEAR/3 lecture |
| explanation |
There is a maximum of three words between online and lecture. You can change this number by changing the number |
| databases |
Web of Science and all Proquest databases such as Sociological Abstracts and Worldwide Political Science Abstracts |
| word order |
In Proquest databases use PRE/x if word order is important. |
ADJx
| example |
online ADJ3 lecture |
| explanation |
There are a maximum of two words between "online" and "lecture" (the last word is included, so you end up with ADJ3). You can change this number by changing the number |
| databases |
Ovid databases such as PsycINFO, Medline and ERIC |
Nx
| example |
online N3 lecture |
| explanation |
There are up to three words between 'online' and 'lecture'. You can change this number by changing the number |
| databases |
EBSCO databases such as Anthropology Plus, Academic Search Premier and CINAHL |
| word order |
If word order is important use W |
Check a database's manual for more information.
... Trunking
If you want to find word variants or both singular and plural, you can replace the end of a search word with a character. This is called truncation. The characters you use for this are called wildcard or joker and are usually an asterisk (*) or a question mark (?).
The effect of trunking depends on where in the word you are trunking.
In the example, you will find far fewer publications with childbirth* than with child*, but they are more specific. Because if your topic is birth, you probably won't be interested in articles about all children.
The character you use to truncate may vary from database to database. Sometimes the number of characters replaced with one character also differs. More information about this can always be found in the help of the relevant database.
Trunking within a word is called masking.
... Masking
Not only at the end, but also within keywords, you can replace one or more letters with a character. This is called masking. The characters you use for this are called wildcard or joker. Usually this is a question mark (?) or a hash (#), but this can vary from database to database. Also, the number of characters replaced by one wildcard or joker can vary from database to database. Check the help file of the relevant database for this.
Masking can be useful if you are in doubt about the spelling of a word or if the spelling may vary.
Examples
|
Globalization
Globalisation
|
Globali?ation
|
|
Behaviour
Behavior
|
Behavio?r
|
|
Colour
Color
|
Col?r |
Globalization
Globalization
Behavior
Behaviour
Color
Color
Old and new spelling: c or k? A?tie
Action Research
Action Research
Between N? Panne?koek
Pancake
Pancake
| Engelse en Amerikaanse spelling |
Globali?ation
Behavio?r
Col?r
|
Globalisation
Globalization
Behavior
Behaviour
Color
Colour
|
| Oude en nieuwe spelling: c of k? |
A?tie |
Actieonderzoek
Aktieonderzoek
|
| Wel of geen tussen N? |
Panne?koek |
Pannekoek
Pannenkoek
|
... Keywords
In library catalogs and databases, you often encounter keywords. In English, these are usually called keywords or subject headings. They are added to the description of a publication to increase the chance that you will find it. For example, if the title of a book does not accurately describe what it is about, a keyword can help you find the book anyway.
Example from CataloguePlus
Conversely, you can also use this knowledge when looking for synonyms and related search terms. If you don't have good search terms yet, but already have a publication that covers the subject matter well, take a look at the publication's keywords in a database so you can use that information to improve your search. Many databases have a keyword list or index where you can find all the keywords used.
Some databases have a thesaurus. In it you can also see the interrelationships between keywords and topics.
... Different search entries
All information about publications in library catalogs and databases is stored in records. Each record has separate fields for title, author, keywords, year of publication, etc. You can search each field separately. The fields are therefore called search entries.
Usually you type your search in the first search bar you come across (simple search or basic search). You search all the fields at once. In full-text files, such as Nexis Uni, you even search the entire text of the document.
This can be useful, but often produces many irrelevant hits. It is better to choose a specific search entry. In databases and catalogs, an advanced or multifield search usually includes a search form where you can easily choose your search entry.
Example Sociological Abstracts
Suppose you are looking for publications on the well-known anthropologist Malinowski:
- in the title ("Swindlers and Liars, the Studies of Malinowski")
- as the author's name ("Bronislaw Malinowski," but also "Nicolas Malinowski")
- in the abstract ("Malinowski's functionalist theory was contrasted with structuralism")
Which is the best option always depends on your research question. In this case, the title search will probably yield the most relevant results, followed by the summary search.
With subject-specific databases, it may be worthwhile to see if there are additional search inputs relevant to your research
... Limitations
Looking for a document in a specific language, from a certain period or in a particular appearance? Limitations can help you with that. You can limit your search result using this kind of data, for example, to:
- peer review or not
- a specific language
- a publication date or period
- material type, for example: journal article, newspaper article, book or audiovisual
Note! The function limits can have different names. Common ones are: limits, refine my results, narrow by and filter. The following database specific limits may also be of interest:
- CataloguePlus: availability of the publication. So online or at a nearby library
- PsycINFO: age groups, research methodology. See also the tutorial Search result refinement
- ERIC: education level. This works the same way as for PsycINFO
- Web of Science and Scopus: filter by subject area after you do a search
... Thesaurus
Many databases have a thesaurus where you can find all the keywords used in that database (see also keywords). You can search here for search terms to supplement your query with. Especially subject-specific databases often have a good thesaurus to help you find search terms.
Example
Suppose you are looking for literature on English language education. Since most of the results you find are about learning English in English-speaking countries, you decide to consult the thesaurus of the educational database ERIC. That one gives you the following suggestions:
If you then click through on one of the suggestions, you are usually shown broader, narrower and related keywords. This functionality is therefore the main difference between a thesaurus and a regular keyword list; in a thesaurus, the keywords have a so-called hierarchical relationship with each other. If you are still in the orientation phase or are in the process of coming up with search terms, this can make a thesaurus especially useful.
If you want to know more about working with the thesaurus, check the help section of the database you want to work with.
SEARCH METHODS
When you start a search, you may first type a few words to see if you can find interesting information quickly. We call this search method quick & dirty.
This often gets you a long way, but there are other ways to search for information.
In this section you will learn some of these methods.
... Citation Tracking
In many publications, both online and in print, you will find a bibliography at the back. In it, the author lists the literature used or makes recommendations for more interesting literature on the topic. This way, you can get on the trail of other useful books or articles. This method is also called backward citation tracking because it allows you to find literature that is older than the article.
In databases such as Web of Science, Scopus and Google Scholar, you can also see who is referencing an article. This is also called forward citation tracking or snowballing because it allows you to find newer articles than the one you have in front of you. This is a useful search strategy if you have found a source that is relevant but too old.
Example
The bibliography of this article from Web of Science contains 43 references, and there are 732 articles in Web of Science that have this article in their bibliography.
... Expand and limit
In devise search terms, you have seen that each search query consists of more components, and you have created a set of appropriate search terms for each component.
Adding components (with AND) usually makes your results more relevant, while adding synonyms and related terms (with OR) makes you find more relevant results. This can have a big effect on your results.
Example from Web of Science
| Search Terms |
Results |
| food AND elderly |
5,125 |
food AND elderly AND "well-being" |
162 |
| eating AND elderly AND social |
172 |
| (food OR eating) AND (elderly OR dementia) AND ("well being" OR "quality of life" OR social) |
583 |
As you can see, a lot of combinations are possible. In practice, it will always be a matter of trying out what works and what does not. Therefore, take a critical look at your results and adjust your search query accordingly.
Expanding search terms
You simply start by typing in one search term. Chances are you will get too many results, many of which are also irrelevant. By adding a second search term, you are already searching more specifically. Keep doing this until you get to a point where the result list looks good at first glance. Then, if it turns out that you have very few articles, see which components it makes sense to add synonyms and related search terms.
Limiting search terms
This search method works the other way around. You start with a large number of search terms that you think should all appear in your search results. You put them all sequentially in the search window. Then look at your results to see if there are any words among them that cause a lot of noise and remove them. If you have very few results, you can also consider not using all the components.
Search History
A fine tool for expanding and narrowing search terms is search history. Using the search history, you can build a search query step-by-step by first working out the individual components and then starting to see what combinations yield.
Example PsycINFO
If you don't see a search history feature, check the help file for the particular database. Keep in mind that not all databases have a search history feature.
...Pearl-growing
Pearl-growing is good to use in the orientation phase when you don't know the right search terms. You start a search and then see if the (list of) results, the Web sites or documents themselves, contain different and perhaps better search terms.
Example from Web of Science:
You found the article below during your search for literature on the social aspects of nutrition in the elderly. As you can see, this article contains many possible search terms tha
In Web of Science and Scopus, you can also sort your results by number of citations. The most cited author on a topic is also likely to be an important author within the relevant field. Use the author name in the Author field to search further for publications.
Improve search results
Too few search results?
If the result of your search is too limited, there could be several reasons.
Did you choose the right source?
Use a different database if necessary.
Is it due to your search query?
Check your search query and adjust it if necessary.
- Formulate your question differently.
- Check if all components are needed.
- It could also be that there is little or no information on your subject and you should choose another subject.
Use different or less specific search terms.
- Check your search terms, for example, have you translated your terms correctly and have you also considered synonyms and spelling variations?
- In a subject-specific database you may need to search with different terms than in a general database.
- Note! Using less specific search terms will increase the amount of hits, but may lower the proportion of relevant hits.
Specify fewer limitations.
- Articles from the last two years may be most appropriate, but perhaps a lot was published on your topic in the years before that.
- Don't limit yourself to Dutch-language publications. Much more is published in English.
Too many search results?
If you find too much non-relevant information, there are several ways to adjust your search:
Did you choose the right source?
If you used a general database, try a subject-specific database.
Is it because of your search query?
Check your search question and adjust it if necessary:
- isn't your research question too broad?
- see if you can add any components in a meaningful way
Use different or more specific search terms.
Narrow your search using exact phrase or a proximity operator.
Specify limitations:
- delimit by certain publication types or by publication date
- choose a particular search entry, such as search by author or title
Summary
In this module, you learned techniques that allow you to search efficiently in your chosen information sources.
You now know how to combine keywords with AND, OR and NOT and include spelling variants through trunking and masking. You have learned how to search with exact expressions and you can make smart use of the various search inputs in a database.
You also learned a number of methods to look up more recent or better literature based on your search result. You do this by using literature references and citations and varying the number of keywords. This all depends, of course, on the capabilities of the various databases.
In any case, being able to perform advanced searches is important!
