Characteristics of good review of literature
literature review can be defined as gathering relevant information on a particular subject of subject
1. Define and delimit the topic. Your review should include a careful definition of your topic, and describe carefully what your topic will not include. For example, you may state that your topic includes attribution retraining, but not locus of control research, and state the reason for this.
2. Review previous reviews. What have previous reviews of the literature concluded? How will your literature review differ?
3. State procedures for obtaining relevant literature. If your review is unbiased, it should include systematic procedures for obtaining relevant literature, procedures that could be replicated by someone else who would then obtain the same literature. This demonstrates that you have attempted to find all relevant research, not just the most easily obtained research, and that you have not been deliberately selective (using only that research that reinforces your own opinion) or inadvertently selective (using certain journals because they are known to you, not realizing they only present one perspective). Search criteria can include:
- computer search of relevant databases (e.g, Psycinfo, ERIC, web of science, Dissertation Abstracts) by topic (list descriptors). In my experience, I have found that computer searches reveal about � to 1/3 of available articles, although they are getting more thorough over the years.
- hand search of relevant journals (e.g., search of specific journals that may be specifically relevant to your topic). Since most journals have only 4 issues per year, this task may not be as time consuming as it may seem!
- "ancestry" search, i.e., search of references in relevant articles, chapters and books, for relevant citations
- "descendant" search, i.e., search Social Science Citation Index, for relevant articles, important authors. SSCI will provide you with a list of articles that have cited these articles or authors
- contact experts in the topic area for suggestions about additional references.
4. Describe common independent and dependent variables. This will show how different studies are interrelated.
5. State criteria for evaluating outcomes (this is most relevant for intervention research). If you say a particular treatment was "effective", what are your criteria for saying so? Author’s conclusion? Statistical significance? Criterion measure? Some other metric of your own?
6. Examine covariation of study outcomes with study characteristics. This is a mouthful, but it means, an educational treatment was effective (outcome), but only with elementary and not secondary students (study characteristics).
7. Support conclusions of the review with data presented in the review. Sometimes authors review an enormous amount of literature, then just conclude what their biases were in the first place! When you make specific conclusions at the end of a review, you should clearly state how and from what data sources you reached your conclusions.
Remember the gaps or limitations in the literature reviewed can directly provide your own research question!! Often, the "Discussion" section of relevant research articles provides implications for future research, which may provide ideas and justification for your own proposed research.
Comments
Post a Comment