Database Search Concepts
1) General keyword searching allows you to
look for your term anywhere in the record.
2) Field searching allows you to look for your
term in a specific section of the record.
3) Placing limits or refining your
search usually involves field searching.
4) Phrase searching allows you to
search for two or more words that are adjacent to each other.
5) The common connectors (OR/AND/NOT) used between
search terms are called Boolean operators.
6)
Proximity searching allows you to search for terms by spatial relationship.
(E.g. Search for terms within the same sentence or paragraph or within
a specific number of words from each other).
7)
Truncation and wild cards allow you to search for variations
on a word or term.
(E.g. Searching for wom*n in Lexis-Nexis finds woman
and women.)
1) General keyword searching allows
you to look for your term anywhere in the record.
The terms you search for in a database are referred to as keywords.
In most databases you are not searching the text of an article; you
are searching information about the article (defined as a
record). When you do a keyword search, typically you
are searching the entire record, rather than a particular section of
the record (defined as a
field).
Terms may appear in parts of the record you haven't anticipated. Your
search term or the subject may turn out to be an author's name or the
name of publisher or appear in a record in some other way than you expected.
For example:
If you are interested in information about the city of London, you might
simply use "London" as your search term. However, in a simple
keyword search, you will get hits for any record where the word "London"
appears, including records of articles by and about individuals named
London (think of American writer Jack London) and any records where
London shows up as the city of publication.
You
will often get false hits--records which have nothing to do with the
topic in which you're interested. That's just part of searching.
2)
Field searching allows you to look for your term in a specific
section of the record.
Field searching (sometimes called segment searching) allows you
to search in particular parts of the record (the fields/segments). It
is more precise than general keyword searching because it allows you
to target your search and hopefully reduce the number of false hits
you receive.
For example:
- When you do an Author,
Title, Library of Congress Subject Headings, Journal Title or Call
Number search in the Basic Search of the library catalog (Lucy2),
you are field searching. In each case, you are specifying which section
of the record you want to search.
- Advanced Search allows
an even greater choice of fields:
- When you specify a Source/Publication/Journal
Name in one of the Database, you are searching in the particular section
of each record which lists the name of the periodical in which an
article originally appeared. This is field searching.
- When you search for terms
in the Title or URL when you use a search engine, you are field searching.
Fields/segments may include
author, title, subject, journal title, call number, URL (Uniform Resource
Locator--a Web adress, e.g. http://hudson2.skidmore.edu) , link, or text,
among others, depending on the particular database.
3) Placing limits or refining your
search usually involves field searching.
Some databases allow you to limit or refine your search.
Usually, by limiting or refining you are including or involving some form
of field searching in your search terms.
For example:
- Limiting to only publications
after 1990 is the same as searching the DATE field for numbers greater
than 1990.
- Refining your search to
include English language publications is the same as searching the
LANGUAGE field for the word "English."
Each database will have its
own format for limiting or refining, but the underlying concept is the
same: you are combining a field search with your keyword search.
4) Phrase searching allows you
to search for two or more words that are adjacent to each other.
Phrase searching allows you to search for multiple keywords adjacent to
each other. You can search for phrases throughout a record or specify
a field. The important thing is that the words be adjacent and that
they be in the order you specify. Often double quotation marks around
the words indicate to the database that you want to search for those words
as a phrase (e.g. "college student"). Another common format
is to choose as a phrase, or some equivalent indicator, on the
search screen.
For example:
- In the LUCY2 catalog you
can choose as a phrase from the pull-down menu on the Advanced
Search screen. You will only get hits for records where your search
terms appear adjacent to each other and in the order you specify.
- In the FirstSearch databases,
you can put double quotation marks around your terms to search for
them as a phrase, e.g. "seasonal affective disorder".
Phrase searching is the most
common form of proximity searching.
5)
The common connectors (OR/AND/NOT)
used between search terms are called Boolean operators.
Named after nineteenth-century English mathematician George Boole, Boolean
logic is the standard system used to search databases. It allows
you to combine variables (in the case of database searching--keywords)
with specific operators: AND, OR , NOT. (Other Boolean operators--IF,
THEN, EXCEPT--seldom, if ever, are used in database searching.)
There are several ways that
the Boolean relationship can be expressed:
- In some databases you
must choose the connectors ("AND," "OR," "NOT")
from a pull-down menu and combine terms on different lines.
- In others you must actually
type out the connectors as part of your search.
- Other symbols may be also
be used to indicate Boolean operators. E.g. In some databases, +
is used instead of AND. (Typically you must place the + in front of
all of the terms.)
- Yet another way to express
these concepts is with phrases. Sometimes you must choose all of
these (for AND) or any of these (for OR), or equivalent
indicators. Using such phrases instead of the traditional operators
is referred to as Implied Boolean Logic.
Of the three operators that
are used in database searching, AND and OR are seen most frequently.
If it is available, be especially careful when using NOT, sometimes
it knocks out records in which you are interested.
You will encounter implied Boolean logic on the Advanced Search screen
in the Lucy2 catalog:.
Below are explanations of each of the operators:
 |
OR (Any
of these) As the image to the left illustrates, when you
use OR to combine keywords, you are searching for records in which
any of these terms appear. You will get a hit for every record
in which one of the terms appears.
Broaden your search by using synonyms.
- advertising OR commercials
- company OR corporation
- drug OR narcotic OR pharmaceutical
- teen OR teenager OR adolescent
|
|
| 
|
AND
(All of these, +)
As the image to the left illustrates, when you use AND to combine
keywords, you are searching for records in which all of these
terms appear. You will get a hit for every record in which all of
the terms appear.
Narrow your search by adding more terms.
- presidential AND election
- +literature +Australian
- press AND censorship AND France
- +film +oscar +1998
|
|
 |
NOT
(AND NOT)
As the image on the left illustrates, when you use NOT to combine
keywords, you are searching for records in which one term appears,
but another term does not appear. (Be careful with NOT. It can cause
unforeseen problems with your results.)
Eliminate potential false hits.
- dolphins NOT miami
- revolution NOT sexual
- nirvana NOT cobain
- buffalo NOT wings
|
|
In the examples above the Boolean operators (OR, AND, NOT) are capitalized.
Databases differ in regards to whether or not you need to capitalize the
operators. In some databases you must capitalize. In others it doesn't
matter. Read the "search help" for each database to find out.
(Databases usually don't disallow capitals, so capitalizing your operators
isn't not a bad habit.)
Combining Boolean Operators
Boolean operators can be
combined to create some rather sophisticated searches. Notice how the
third and fourth searches in the chart below provide very different
results than the first two simple searches, just by adding one additional
term..
6) Proximity searching allows you to
search for terms within a specified number of words from each other.
Proximity searching allows you to search for keywords by how close they
are to one another.
The most common form of proximity searching is phrase searching
(see No. 4 above). When phrase searching,
you seek two or more words which are adjacent to each other. In many databases,
phrase searching is the only type of proximity searching available. However,
some databases allow you to search by specifying the number of words away
from each other your keywords must be or specifying that they must be
in the same sentence or paragraph.
In a sense, a proximity search is like "Boolean AND plus"--all
of your search terms must be in the record, plus they must be a
certain distance from one another or must be together in a specific part
of the record.
Proximity searching is usually only available in full-text databases.
It is most effective when used to search text rather than citations or
even abstracts.
- In Lexis-Nexis Academic
Universe, you can specify that search terms be in the same sentence,
the same paragraph, or within 5, 10 or 25 words from another.
Here are some examples taken from Academic Universe's Search Tips:
w/sentence
- Looks for documents with search words in the same sentence.
Example: doctor w/sentence malpractice finds doctor in the
same sentence as malpractice.
w/paragraph - Looks for documents with search words in the
same paragraph.
Example: doctor w/paragraph malpractice finds doctor in the
same paragraph as malpractice.
w/n - Links search words and phrases to create concepts.
You can choose within 5, 10 or 25 words. W/n does not specify a
word order.
Example: doctor w/5 malpractice finds doctor within 5 or fewer words
of malpractice, regardless of which word appears first.
Each database that allows
proximity searching has its own format for specifying it. (FirstSearch
uses a somewhat different format for doing proximity searching.) Consult
the search help in each database for the proper search terms.
In some databases, the order of the search terms is specified; in others,
it doesn't matter which word comes first.
7) Truncation
and wild cards allow you to search for
variations on a word or term.
Some databases allow you to use truncation and/or wild cards.
These allow you to search for variants of a word by using a symbol
to represent one or more characters. The most common symbols are * (asterisks),
? (question marks), ! (exclamation marks), + (plus signs). Truncation
usually refers to searching for variant endings and wild cards
usually refer to symbols used to to look for variants within a
keyword, but sometimes these terms are used interchangeably.
Truncation usually refers to searching
for variant endings. Some databases automatically search for variant
endings; this is called automatic truncation. You don't need to use a
truncation symbol in databases that automatically truncate.
In databases which offer
truncation, but don't automatically truncate, a truncation symbol is
used:
- In FirstSearch, a + is
used to truncate for plurals.
car+ finds car, cars, car's and cars'
If truncation is not supported
by the database's search engine use the Boolean operator OR
to combine like terms.
Wild Cards usually refer to symbols
used within a keyword to search for variants.
- wom*n finds woman and
women
- goldst**n finds goldstein
and goldstien
Remember: Each database uses
particular symbols. Consult the search help for the database you are using.
If you have questions
about database search concepts,
ask the Reference Librarian, tel (518) 580-5503.
|
| |
|
| Back to Database Searching |
|
|
