Conducting a literature review involves using research databases to identify materials that cover or are related in some sense to the research topic. In some cases the research topic may be so original in its scope that no one has done anything exactly like it, so research that is at least similar or related will provide source material for the literature review. The selection of databases will be driven by the subject matter and the scope of the project.
Selecting Databases -- Most academic libraries now provide access to a majority of their databases and their catalog via a so-called discovery tool. A discovery tool makes searching library systems more "Google-like" in that even the simplest of queries can be entered and results retrieved. However, many times the results are also "Google-like" in the sheer quantity of items retrieved. While a discovery tool can be invaluable for quickly finding a multitude of resources on nearly any topic, there are a number of considerations a researcher should keep in mind when using a discovery tool, especially for the researcher who is attempting a comprehensive literature review.
No discovery tool works with every database subscribed to by a library. Some libraries might subscribe to two or three hundred different research databases covering a large number of subject areas. Competing discovery systems might negotiate agreements with different database vendors in order to provide access to a large range of materials. There will be other vendors with whom agreements are not forthcoming, therefore their materials are not included in the discovery tool results. While this might be of only minor concern for a researcher looking to do a fairly limited research project, the researcher looking to do a comprehensive review of the literature in preparation for writing a master's thesis or a doctoral dissertation will run the risk of missing some materials by limiting the search just to a particular library's discovery system. If only one system covered everything that a researcher could possibly need, libraries would have no need to subscribe to hundreds of different databases. The reality is that no one tool does it all. Not even Google Scholar.
Book collections might be excluded from results delivered by a discovery tool. While many libraries are making results from their own catalogs available via their discovery tools, they might not cover books that are discoverable from other library collections, thus making a search of book collections incomplete. Most libraries subscribe to an international database of library catalogs known as WorldCat. This database will provide comprehensive coverage of books, media, and other physical library materials available in libraries worldwide.
Features available in a particular database might not be available in a discovery tool. Keep in mind that a discovery tool is a search system that enables searching across content from numerous individual databases. An individual database might have search features that cannot be provided through a discovery tool, since the discovery tool is designed to accommodate a large number of systems with a single search. For example, the nursing database CINAHL includes the ability to limit a search to specific practice areas, to limit to evidence-based practice, to limit to gender, and to search using medical subject headings, among other things, all specialized facets that are not available in a discovery tool. To have these advanced capabilities, a researcher would need to go directly to CINAHL and search it natively.
Some discovery tools are set, by default, to limit search results to those items directly available through a particular library's collections. While many researchers will be most concerned with what is immediately available to them at their own library, a researcher concerned with finding everything that has been done on a particular topic will need to go beyond what's available at his or her home library and include materials that are available elsewhere. Master's and doctoral candidates should take care to notice if their library's discovery tool automatically limits to available materials and broaden the scope to include ALL materials, not just those available.
With the foregoing in mind, a researcher might start a search by using the library's discovery tool and then follow up by reviewing which databases have been included in the search and, more importantly, which databases have not been included. Most libraries will facilitate locating its individual databases through a subject arrangement of some kind. Once those databases that are not discoverable have been identified, the researcher would do well to search them individually to find out if other materials can be identified outside of the discovery tool. One additional tool that a doctoral researcher should of necessity include in a search is ISI's Web of Knowledge. The two major systems searchable within ISI's Web are the Social Sciences Citation Index and the Science Citation Index. The purpose of these two systems is to enable a researcher to determine what research has been cited over the years by any number of researchers and how many times it has been cited.
Formulating an Effective Search Strategy -- Key to performing an effective literature review is selecting search terms that will effectively identify materials that are relevant to the research topic. An initial strategy for selecting search terminology might be to list all possible relevant terms and their synonyms in order to have a working vocabulary for use in the research databases. While an individual subject database will likely use a "controlled vocabulary" to index articles and other materials that are included in the database, the same vocabulary might not be as effective in a database that focuses on a different subject area. For example, terminology that is used frequently in psychological literature might not be as effective in searching a human resources management database. Brainstorming the topic before launching into a search will help a researcher arrive at a good working vocabulary to use when probing the databases for relevant literature.
As materials are identified with the initial search, the researcher will want to keep track of other terminology that could be of use in performing additional searches. Sometimes the most effective search terminology can be found by reading the abstracts of relevant materials located through a library's research databases. For example, an initial search on the concept of "mainstreaming" might lead the researcher to articles that discuss mainstreaming but which also look into the concept of "inclusion" in education. While the terms mainstreaming and inclusion are sometimes used synonymously, they really embody two different approaches to working with students having special needs. Abstracts of articles located in the initial search on mainstreaming will uncover related concepts such as inclusion and help a researcher develop a better, more effective vocabulary for fleshing out the literature review.
In addition to searching using key concepts aligned with the research topic, a researcher likely also will want to search for additional materials produced by key authors who are identified in the initial searches. As a researcher reviews items retrieved in the initial stages of the survey, he or she will begin to notice certain authors coming up over and over in relation to the topic. To make sure that no stone is left unturned, it would be advisable to search the available, relevant library databases for other materials by those key authors, just to make sure something of importance has not been missed. A review of the reference lists for each of the items identified in the search will also help to identify key literature that should be reviewed.
Locating the Materials and Composing the Review -- In many cases the items identified through the library's databases will also be available online through the same or related databases. This, however, is not always the case. When materials are not available online, the researcher should check the library's physical collections (print, media, etc.) to determine if the items are available in the library, itself. For those materials not physically available in the home library, the researcher will use interlibrary loan to procure copies from other libraries or services. While abstracts are extremely useful in identifying the right types of materials, they are no substitute for the actual items, themselves. The thorough researcher will make sure that all the key literature has been retrieved and read thoroughly before proceeding too far with the original research.
The end result of the literature review is a discussion of the central themes in the research and an overview of the significant studies located by the researcher. This discussion serves as the lead section of a paper or article that reports the findings of an original research study and sets the stage for presentation of the original study by providing a review of research that has been conducted prior to the current study. As the researcher conducts his or her own study, other relevant materials might enter into the professional literature. It is the researcher's responsibility to update the literature review with newly released information prior to completing his or her own study.
Updating the Initial Search -- Most research projects will take place over a period of time and are not completed in the short term. Especially in the case of master's and doctoral projects, the research process might take a year or several years to complete. During this time, it will be important for the researcher to periodically review the research that has been going on at the same time as his or her own research. Revisiting the search strategies employed in the initial pass of the ltierature will turn up any new studies that might have come to light since the initial search. Fortunately, most research databases and discovery systems provide researchers with the means for automatically notifying them when new materials matching the search strategy have entered the system. This requires that a researcher sign up for a personal "account" with the database in order to save his or her searches and set up "alerts" when new materials come online. Setting up an account does not involve charges to the researcher; this is all a part of the cost borne by the home library in providing access to the databases.
