As recently as a few years ago, if you wanted to be considered a cutting-edge expert in computer-assisted-legal-research (CALR) you only had to learn to use Lexis and Westlaw. While Lexis and Westlaw are still the major CALR vendors, you must be able to make use of other electronic databases as well as the Internet to do minimally competent research.
The general theory of electronic research comes in twos. Electronic information resources come in two basic levels of completeness: either full text records or abstract/index records. There are two basic linguistic difficulties in finding information: synonymy and ambiguity. There are two basic approaches for overcoming those difficulties in finding information: either subject or keyword search methods. There are two basic measures of search efficiency: precision and recall. And there are two basic tools for improving search efficiency: Boolean searching and field searching.
Electronic resources have varying levels of completeness. Some resources are full-text databases, for example, newspaper articles on the Baltimore Sun’s web site at http://www.baltimoresun.com. Other resources are abstracts or indexes; a common example is an online library catalog. This type of resource does not provide the full text of the item, but provides varying levels of detail about the item, and citations for retrieving the full text. Some vendors provide databases of both types -- Lexis and Westlaw both have databases that are full-text and databases that are indexes only. Databases available on the Thurgood Marshall Law Library’s web page come in different varieties: full-text, partial full-text, and index/abstract only. Some resources that have full text may only have that text for a limited period of time. Beyond that time frame you may have to pay a document retrieval fee or use a subscription database to retrieve articles. The level of completeness is a factor to consider when you choose a certain resource.
Databases also come with varying degrees of structure. For example, a library’s catalog has limited records, but the information is in a highly structured format that makes it easy to search for items by subject heading, author, or title. The Internet can be considered a database with virtually unlimited records and with virtually no structure. The nature of the database dictates the methods that should be used to most efficiently search it. One of the first steps in successful electronic research is selecting the appropriate database and learning enough about the database to pick an efficient search method.
In all research the searcher combats some basic linguistic difficulties: ambiguity and synonymy. Electronic research, especially full text, compounds these difficulties. When searching, you are normally looking for a concept, not a particular set of words. But one of the major tenets of “good” writing is to vary your words and find creative, new ways of expressing your ideas. Synonymy is the problem of there being many ways to express a single concept. For example, “sentenced to die,” “death penalty,” and “capital punishment” all express essentially the same concept. Probably you are in law school because you want to become something that can be called an attorney, a barrister, counselor, lawyer, litigator, or solicitor. Electronic research deals only with words, not concepts. To do comprehensive research, the searcher must account for the possible synonyms for the concept being sought. The flip side of the problem is ambiguity. Ambiguity is the problem of multiple concepts all being expressed by the same words. If you were to search one of the case law databases for “release” you would find criminal cases where the sentence was life without release, cases where a plaintiff signed a liability release, and cases about faulty auto brake releases. A counselor could mean an attorney, but it could also mean a social worker; a solicitor could mean an attorney, but it could also mean a salesperson. Overcoming synonymy and ambiguity is a first step in effective research.
There are two ways to measure effectiveness of your search. The first measure is precision, which means finding only what you want to find. A precise search for documents about attorneys would include the documents where “counselor” means attorney but exclude all the documents where it meant a social worker. The second measure is recall. Essentially this means finding everything that you want to find. A search for documents about attorneys with good recall would find articles that use any of the words barrister, counselor, lawyer, or litigator. No one but the searcher can determine how precise a search needs to be or how comprehensive the search needs to be. At times a quick and dirty search that retrieves a few examples is all that is needed. At other times the research has to be as comprehensive as possible to retrieve every single case or item on point.
Two major tools exist in electronic research for increasing search efficiency. The first is Boolean searching. You have most likely used this at least occasionally when doing research on the Internet. This tool is also available in a much more sophisticated fashion on Lexis and Westlaw and on many other electronic resources. The second is field searching. This technique takes advantage of the structure of the database. It is available to a degree on the Internet, but is much more helpful in the more highly structured databases like Westlaw and Lexis.
There are two basic ways you can approach finding information in electronic format: a subject approach or a keyword approach. A common example of a subject approach is the old library card catalog. (Modern online library catalogs offer both the option to search by subject and the option to search by keyword.) A classic example on the Internet is YahooÔ http://www.yahoo.com/, the directory approach. When you search by subject, you search subject headings that have been assigned to an item as descriptive of the item’s contents. Another example of the subject approach specific to legal research is the West Topic and Key Number system. The important thing to remember about the subject approach is that someone (or even a program) has analyzed the item and assigned a subject descriptor to it. This means that some of the work is done for you. For example, by doing a subject search for the heading “homicide” you would find items dealing with that subject even if the item uses the words “murder,” “stab,” “kill,” or “fatally wound” without using the term homicide. Also, a subject search for the heading “homicide” would not retrieve an item about the filming of Homicide: Life on the Street. The subject approach is one way to compensate for the problems of synonymy and ambiguity. While some of the work is done for you, subject searching also means that you are depending on the work having been done correctly, and you are limited to subject headings assigned by somebody else.
When you search using the keyword approach, you are searching for any appearance of your search term. In full text databases this method searches for your term in the title field of the item, the text field, the assigned subject headings, and any other field in the item. The important thing to remember about the keyword approach is that it searches for character strings only, and does not search for concepts. This means that for whatever information you seek, you must decide what terms best describe it. Keyword searching is particularly helpful when the concept you seek is very new and has not yet had a chance to be incorporated into subject headings.