Reviewing the Literature

Undertaking a review of the literature – and crafting a concise and informative literature review – often presents a major challenge for researchers.  Many people are surprised by their hesitation at completing this crucial task; after all, most of us had experience reviewing literature for papers written in high school or college.  The literature review portion of a thesis/dissertation or research article serves a very different purpose, however, than the type of literature review we may have previously constructed (which usually consisted of a series of descriptions of what was reported in secondary sources).  The literature review portion of scholarly work moves beyond simply reporting on past research and instead must synthesize existing research in a way that demonstrates the need for additional research in that area.

Where to Start

Sorting through the vast literature available on most topics can seem like an endless task, and it can be endless if allowed to be.  Researchers often feel that they must review every single article that is even remotely related to their topic; this is especially true when exploring a new topic area, as researchers fear that their lack of expertise in a particular area may lead to an embarrassing omission in their review of the literature.  Fighting the urge to read everything requires a great deal of discipline, for without a clear strategy is it easy to spend months (or even years!) reviewing the literature without actually producing a usable literature review.

One commonly-overlooked resource for literature reviews are secondary sources.  Secondary sources are literature reviews prepared by other researchers; these are often published in journals that specialize in summarizing the literature in certain fields (examples include the Review of Research in Education or Psychological Bulletin).  There are also a number of books and handbooks (often published at the end of each year) which feature literature reviews.  Secondary sources offer a good overview of particular topics and can reveal what is already known.  Looking at the references for such reviews can provide a starting point for identifying the central literature in the area.  Such reviews can also help to identify key words for use in searching databases.

Database searches form the basis of most searches for literature.  Most disciplines have their own well-known databases, but it is also worth searching related databases with more general coverage (e.g., Web of Science, PubMed, ERIC).  Google Scholar is becoming more refined and has the added benefit of showing the number of citations each article has received, which can help identify the most important articles in a particular field.

 

Sorting through Articles

Once a handful of pertinent articles have been located the process tends to take off, as each reference list provides countless additional articles to review.  This is the point at which discipline is most needed – remember that you cannot read everything!

Following are some general guidelines for identifying the most important articles: More recent work is preferred over older (with the exception of key original works which ground the field; these will become apparent through repeated citations in contemporary articles).  Work published in more prestigious and main-stream journals is preferred over that found in lesser-known outlets (examining the impact factor of journals or asking colleagues/faculty can help identify these sources).  Rigorous empirical work is generally preferred over opinion pieces, though such work can be useful for developing new ideas.

Reviewing Articles

In addition to recording basic information in an annotated bibliography format, it is useful to also reflect on what each article adds to your thinking about the topic (similar to memoing when conducting qualitative research).  It can be helpful to use a consistent format for each review, including the following items: Title, Authors, Main Ideas Presented in Lit Review, General Research Question, Specific Hypotheses, Sampling Design/Population of Inference, Operationalization of Key Variables, Statistical Techniques, Key Findings, Limitations/Future Research, Applicability to Current Research Project, Additional References.

A template for all article reviews can be created (in Word or Excel), though reference management software such as EndNote or ProCite is an invaluable tool that should be seriously considered by anyone conducting a literature review (and especially those facing comprehensive exams).

Rationale for the Study

While a literature review offers a general overview of the existing literature on a topic, the main purpose of a literature review in a scholarly work is to provide the rationale or reason for the current study.  As the vast majority of research builds on existing ideas, it is essential that researchers demonstrate the gap the proposed research seeks to address.  Usually the motivation for new research derives directly from the process of reviewing the literature, during which time one or more of the following may arise:

Inconsistent results: Often when reviewing the literature it becomes clear that the results from existing studies are inconsistent or conflicting.  The design of the studies or the purposes of the researchers may account for the observed differences, but often these conflicts point to questions that are in need of additional investigation.

Problems in previous research designs: As research methods (and methodological sophistication) continue to develop, the opportunity to conduct more rigorous research arises in many areas.  In reading the literature on a particular topic it may become clear that the majority of existing studies have relied solely on one perspective (quantitative or qualitative), have all been cross-sectional in nature, or have relied on observational methods rather than controlled designs.  Mixed-methods approaches, longitudinal studies, and carefully designed quasi-experimental studies can address each of these concerns, respectively.  With the continuing development of the Internet it is also becoming possible to locate previously hidden populations for inclusion in studies, which represents another correction to previous research designs.  Indeed, including the experiences of previously understudied populations is a common source of new research opportunities.