A summary of the source, normally a paragraph or two, that is provided by the author when submitting for publication. A key field to look within for relevant terms.
The classic ones are AND, OR and NOT; these are used to combine keywords or search strings in particular ways. There are additional terms like proximity operators that combine things in specialised ways. Named after George Boole, a professor of mathematics who taught in Cork!
Most databases have a list of standardised terms for subjects called subject headings; everything on that topic can then hopefully be found under the relevant heading, if one exists. Individual sources are tagged by the database to be about those concepts, no matter what words were used in the source itself. Often called a thesaurus in a database.
An indexed area in a source (e.g., a journal article or book chapter) such as the title, year of publication, and author. There are normally lots of these for any record, some derived from the source itself, some added by the database after studying the source.
Can refer to the author's keywords – which are supplied by the author and may be a field you can search - or just the terms you want to use in your search. However, some databases use the term "keywords" specifically for their controlled vocabulary or other topic headings, so it's a bit confusing.
Also called filters. These are restrictions on your search that can be applied by making selections within the database, such as stating the articles need to be in English, or published after a certain date.
MeSH stands for Medical Subject Heading. These terms are a type of controlled vocabulary in biomedical databases like MEDLINE and PubMed, as well as several other medical databases that use the exact same list of subject headings. Sometimes used as a synonym by people for controlled vocabulary in other databases that aren’t the same list, so watch out.
A way of specifying the order terms are combined using Boolean operators. This allows you to know exactly what is being combined together in clauses, so that the correct operations are applied and it does what you think it is doing.
Some examples:
(Cat AND Dog) OR Rabbit
...means it must mention both Cat and Dog, or, alternatively, could just mention Rabbit.
(Cat) AND (Dog OR Rabbit)
...means it must mention Cat, but then also either Dog or Rabbit (or indeed both).
But!
Cat AND Dog OR Rabbit
...without any brackets can be interpreted in those two different ways, depending on the database and in what order it applies the operators.
Nearly all databases let you search for phrases – those exact words in that exact order – using quote marks. Some databases have an issue with the curved Smart Quotes that Words uses automatically, so it’s better to stick to simple quotes "like these ones" - you may need to turn off Smart Quotes in Word. If you need a few words in-between your search terms, you can explore proximity operators.
People often confuse the name of the company that makes the database, or the platform that it's on, with the database itself. For example, EBSCO is a company that provide access to many databases. There is no database called "EBSCO", however. The name of its platform - EBSCOhost - implies there is a commonality in the ways the databases look and feel, and the syntax they use to run searches. Again, there is no database as such called "EBSCOhost".
It's often possible to search multiple databases at the same time on one platform; however, while this is tempting, it's not always the best approach!
Some of those databases (ERIC, Medline, PsycInfo, for example) from EBSCO are also available on other platforms from other providers, such as Ovid and ProQuest. They look and work differently to each other, but the same database from another company should have the same data within it. Sometimes, for legacy reasons, a provider can have different platforms that give access to multiple databases that work in different ways - for example, when two companies have merged.
These are a special type of combining tool that mean words have to be close to each other within a field, rather than being anywhere in that field or indeed in the record. Sometimes you can specify how many words there can be in-between, or what order they have to be. The syntax for this will vary from database to database; PubMed doesn’t allow this, but most others do.
A search normally looks for a number of terms. When put together this is referred to as a search string. This can refer to the whole search put together that can be pasted into a database and run as-is, or individual elements within it (so, strings or strands). May include the syntax to make it look in different fields, or may just be the important words.
Synonyms of your terms are a vital part of a search, generally used in fields like the title and abstract. You needs to think of words that mean the same thing, in case someone uses that in the source instead of the word you originally thought of. So for "cancer*" you might also use "oncolog*", or "tumour*" OR "tumor*", or "neoplasm*", or "malignan*", or "carcinoma*" (note all the wildcards!). Plus the names of different types of cancers, like leukemia. These are added together using the Boolean operator OR.
Codes and words in a search that tell it to look in particular fields (like the abstract or title index) or to combine searches in different ways using Boolean operators. Different databases are likely to have wildly different syntax, but ones provided by the same company tend to use the same codes, if they are on tghe same platform.
Wildcards are a way you can search for variations of a word, such as plurals, verb forms and UK/US spellings. How you do this depends on the database, but nearly all will let you use a stem word with an asterisk to find longer variants. Some databases allow you to have variant letters in the middle of the word, but that’s often trickier to use.
