Module 3: Searching

Objectives:

• Distinguish a database from other types of information collections
• Identify the coverage of a database
• Understand the concept of field, or advanced searching in a database
• Construct effective search queries using logical operators and related strategies

This quality of being " ... arranged for ease and speed of search and retrieval" is what distinguishes a database from a computer network like the Internet, which has no standardized organization principle.

Databases may sometimes be accessed through the Internet, but their contents are not retrieved by search engine services like Google or Yahoo! Most are available through separate Web sites that charge a fee for use, normally paid by libraries on behalf of their users.

Every database contains only certain types and amounts of information, a characteristic called coverage. This information can typically be found in the database itself under links such as "About [name of database]," "Database information," "Title list," or "Sources," etc. Web-based databases are typically accessed from a link that is annotated with some information about coverage. Databases published in paper form normally locate this information in the front of each volume or in an introduction.

Consider the following elements of database coverage:

Elements	Example
What kinds of documents?	Journals, magazines, books, book chapters, dissertations, audio files, statistical tables, images, Web pages, software applications?
Which disciplines?	Sociology, music, chemistry, all, none?
What time periods?	The current year? 1960-1998? How often is the database updated? Hourly, daily, weekly, monthly, annually?
What languages?	English only? Other languages?
Which publication types?	Scholarly? Popular? Trade? All three? Others?
What is included in the record?	A whole article or chapter (full-text) or just a brief description (bibliographic citation and abstract)? Publisher and title?

It is unrealistic, at this time, to expect all information sources to come directly to a computer screen on demand. The reasons for this will be clearer to you as you become more experienced finding and using information. In the meantime, having a record describing a source that you can then find using your own abilities and knowledge is an excellent starting point. The description in a record uses elements called metadata [information about information]. In many cases, the text itself will have to be located using other finding tools. Some libraries provide links to the text of articles, when available, at their own expense.

For example, your driver's license or ID card is a record about you with fields describing your name, eye color, height, address, and so on. Field names can vary, but in library databases author, title, subject, publisher, and publication date are common.

Most search engines and databases search "words anywhere" or "keywords" automatically unless you select another type of search.

Keyword searching finds matches for your terms in any field of a record or any part of a Web page, so you will typically retrieve more information with less precision. This is known as "recall" searching because it focuses on recalling as much information as possible.

Databases and search engines may allow searching in specific fields such as author, title, url (Web address), or subject and will sometimes refer to this as "advanced," or "expert" searching. These searches will typically retrieve less information with more precision. This is called "precision" searching because it focuses on finding only precisely what you need.

For example:

RED + BLUE + GREEN		Tie=RED, Shirt=BLUE, Beret=GREEN
If you are a detective and the only clues you have for a missing persons case are the words "red," "blue," and "green," these people could be a match. This is keyword searching.		If instead you knew your person had a red tie, blue shirt, and a green beret, you have a better chance of finding the right person. This is field searching.

Phrases | logical operators | truncation | wild cards | nesting

Most databases don't understand the natural language we speak and need help understanding what we're looking for. For this, they require a special set of conventions, including:

Conventions		Description
Quotation marks		Around exact phrases (e.g. "university of washington")
Logical or Boolean operators		Connecting words that narrow or broaden a search to include only what you need. Examples: OR, AND, NOT
Wildcards and truncation symbols (* # ? !)		For terms that have variant forms of spelling or different possible endings. Examples: child* for child, children, childhood, childish, etc.
Nesting		Placing terms in parentheses to indicate separate units. (Like an equation, (A or B) not C

Databases and search engines apply these rules differently, so check HELP files to find out how to use them.

Click on the links below for a demonstration of each strategy. Select either an animated flash or a static image. They all pop-up in new windows.

Function		Search Strategy		Animation		Image
Narrowing		AND NOT
Broadening		OR Wildcard
Combining		Nesting

Here are some tips if you found too much information, too little information, or the wrong information in your search.

Too Much Information

1. Try looking at an irrelevant record your search retrieved.
   Can you figure out why the database gave it to you? Did you use one word that the computer misunderstood? See if you can use a more specific term or maybe a short phrase that excludes the meaning you don't want. Try adding a new term which makes your old term more specific.

   Instead of Japan and economy
   
   Try  Japan and economy and (auto or automobile or car)

2. Check where in the record your search terms matched.
   The best matches for topics are in fields like Subject or Title . Look for an Advanced or Expert Search option in the database to search in specific fields only, if you can.
3. Use limiters when they're available.
   Will the database let you ask for publications only in English? Can you ask for only journal articles? Want more recent information? Is there a subject heading that covers your topic? Can you get rid of book and film reviews? Play around with your options and see if they help. Try using the operator NOT.
    

   Examples
   (Iran and Iraq)	not		war
   Hussein	not		Saddam
   Clinton	not		Lewinsky
   +Jazz -Utah

Too Little Information

1. Did you spell your search terms correctly?
   Research databases are remarkable tools, but they don't come equipped with spell checkers. One misspelled word can sink an entire search. Check a dictionary.
2. Get rid of long phrases.
   When you type in a phrase, all the words must appear in exactly that order before the database will give you anything. Some databases automatically put the operator AND between the words you type, turning your phrase into a long Boolean search string.

   

3. Try using alternative terms.
   That's what you gathered all the extra vocabulary for. Don't forget truncation or wildcards for variant forms of a word.
4. Try to come up with broader terms for the idea you need.
   Every so often, it happens that there's very little written on a specific topic, but a lot on the general area.
    

   Examples
   Very Narrow		recombinant DNA and sheep
   Narrow		cloning and animals
   Broader		genetic engineering and animal*
   Very Broad		genetic* and animal*

The Wrong Information

1. Check the coverage of the databases you're using. Do they cover the kinds of material you need? The right discipline(s)? The right kinds of documents? The right dates?
2. Try looking up databases by subject. Click on Find Articles, then use the drop down menu to choose a broad subject area. Click on the INFO button for each database to see its coverage.
3. Try using the Journal List (Serials solution) link. You have to know what general field your subject falls under (Social and Behavioral Sciences? Business and Economics? Fine Arts and Music?) Try a few of these and see where you can find your subject. Then try using some of the databases you find linked there.