Searching Databases
Searching for periodical literature has come a long way since the days of using the old Reader's Guide to Periodical Literature. Rather than use printed indexes to periodical literature, many researchers find that electronic alternatives provide them not only with subject and author access to periodical articles, but, in many cases, also with brief summaries of the articles. Electronic indexes offer many advantages over the print indexes, the most notable being:
- increased searching flexibility (use of boolean searching to find articles by keywords, subjects, authors, etc., and the ability to search ALL indexed fields, including abstracts, and even full text)
- frequent updates
- improved accessibility (multiple users can access the same index simultaneously as long as the library has bought sufficient licenses)
- fast search results
- customizable results
- printing, downloading, and emailing capabilities
Many databases also include the full text of many of the indexed articles. This creates for the researcher a kind of "one stop shopping" for research materials.
Currently, the UNF Library offers database searching primarily through online systems. For example, FirstSearch, CSA, Gale/Infotrac, ProQuest, Science Direct, and Lexis-Nexis systems are all online/Internet based search systems that allow users to access hundreds of subject databases.
The purpose of databases is to provide researchers with a flexible way of finding information within numerous publications (typically periodicals) without having to search each publication individually using its index or table of contents. A database frequently includes hundreds or even thousands of publications within its scope and provides more detailed access to information than a printed index could possibly provide. For example, the Readers' Guide arranges articles by subject heading and by author and has separate listings of book reviews. Finding an article by a certain author on a certain subject within a particular magazine or journal would be very tedious and time-consuming in the print version of the Readers' Guide. The online database version, known as the Readers' Guide Abstracts, provides access by subject, by title, by journal title, by author, by any number of means, including access to abstracts. What's more, researchers can search multiple criteria in one search, thus making finding an article by a certain author, on a certain subject, within a certain magazine or journal, really quite simple.
The key to the flexibility offered by databases is in how they are designed. Most provide access to article descriptions based on specific database fields, assigned to each article. For example, we normally would expect to find information on an article that would include the article title, the article author, the name of the magazine or journal, the date the article was published, and the page numbers on which the article was printed. All of these data are indexed in a database in individual fields, so each bit of information can be searched directly. Additionally, more detail is provided for each article, including subject description and even abstracting. The result of this field-oriented design in databases is that researchers can perform fairly sophisticated searching in a short time.
Typical Database Item Description Format
Items within a database are organized into fields, each of which can be searched directly. For example, a typical database record might have the following field arrangement.
| TI | Confusion and Fear in the World of Database Searching |
| AU | Addled, Amos |
| IN | Institute for the Natural Study of Adult Nervousness with Electronics (INSANE) |
| SO | Journal of Electronic Disorders, 19(1) 2001, pp. 60-75 |
| IS | 0083-3001x |
| PB | Kalamazoo, MI: INSANE, http://www.insane.org |
| AB | Recent research suggests that library users who have a beer prior to using a database are less likely to fear using an online database. |
| LA | English |
| PY | 2001 |
| PT | Peer Reviewed Journal |
| PO | Human |
| DE | *databases; *phobias;*computer literacy; *research; *case studies |
| DE | alcohol and computers; fear |
| UD | 0103 |
| AN | 2001-3465 |
This format can vary from database to database, including more or sometimes less information for each item indexed. Other database suppliers may use different approaches for presenting field indexing, but most follow a format in many ways similar to the above example.
In the foregoing, the abbreviations can be interpreted as follows:
| TI | Article title. | |
| AU | Author's or Authors' name(s). | |
| IN | Institution (organization with which the author is affiliated or which sponsored the research). | |
| SO | Source publication, which is typically the journal or magazine which published the article. This information typically includes the volume, issue number, issue date, and pages of the article. | |
| IS | ISSN (International Standard Serial Number) -- A unique identifying number for the source publication. | |
| PB | Publisher (this may or may not be included, depending on the type of publication and the database). | |
| AB | Abstract -- a brief summary of the article contents. | |
| LA | Language in which the article is written (many databases may included non-English language articles). | |
| PY | Publication Year. | |
| PT | Publication type (this may also be listed as DT, Document Type). | |
| PO | Population -- often used in research databases to describe the research population by type (human/animal), gender (male/female), and by age grouping. | |
| DE | Descriptor -- or, more simply, subject (descriptors are assigned to help the researcher better judge the article content based on a set or controlled vocabulary). Major descriptors will often be listed with asterisks immediately following. | |
| DE | Descriptor -- further, minor descriptors. | |
| UD | Update. Many databases provide a date or code that indicates when the information in the database was last refreshed. | |
| AN | Accession Number -- as articles are received for inclusion in a database they are typically assigned sequential numbers as a means of tracking when they were received and in what order. |
The Value of Indexing and Searching by Field
Most electronic databases are normally searchable by keyword, subject, author, and title using Boolean search operators AND, OR, and NOT and proximity operators such as adjacent, near, and within. The ability to limit a search to a specific bit of information about an article can be extremely powerful and can save a researcher hours. A simple example should help demonstrate.
A researcher has been told by a professor that author Peter Drucker wrote an article on the effects of technological change on organizational management sometime during the 1980s. The professor doesn't remember which journal actually carried the article. The researcher decides to go to the library and to use one of the business databases to find the article.
Choosing the ABI/Inform database, the researcher begins the search by typing in the basic keyword search "peter drucker." The basic search delivers a result listing 3,397 articles. Deciding that browsing several thousand articles is not a very effective way to find the article recommended by the professor, the researcher types into the basic search entry box the words "peter drucker technological change," thinking that this should do the trick. The result is much better, but there are still 38 articles listed for this search, and none of the articles listed on the first result screen is even by Drucker. What to do?
The solution that provided the best results was to go to an "advanced" search screen in ABI/Inform and to specifically search for articles by Peter Drucker and to search the article abstracts for the phrase "technological change." The researcher also limited the search to the time period from 1980 through 1989. The Boolean syntax is automatically built into the advanced search screen: the default connector is AND. The result: 5 articles, all by Peter Drucker, that somehow addressed technological change and management. After reviewing the article titles with the professor, the researcher found that the article "Managing for Tomorrow--Managing in Turbulent Times" from the journal Industry Week was the one the professor had in mind.
While this search scenario had a fairly quick resolution, in many cases a researcher will not have as many details to go on and may be faced with looking through hundreds of articles to try to find what he or she needs. That is, if basic search is the only approach being used. The ability to limit to specific fields (author, title, journal, abstract, etc.) provides the researcher with precise control of the results returned by the database and can save literaly hours of time.
What This All Means
Databases are very highly organized systems of information that can be queried very simply or can be manipulated to pinpoint with a high degree of precision very specific information. As the organized systems that they are, databases are also predictable and therefore fairly easy to learn to use. The wise researcher will spend a bit of time learning how the database is organized and how to search effectively before launching into a major research project. Basic keyword searching is typically the sloppiest possible way to find information; the better search strategy takes advantage of the field indexing provided by the database.
Page updated 01/07.
