structure of a literature review

structure of a literature review

Importance of a Well-Structured Literature Review

1. Introduction

After examining various thesis analyses, it is clear that almost all of them have just a short section dedicated to the literature review. Most of them would only say, “The review of the literature supported the problem statement and the currently proposed solution algorithm. The results presented in this chapter have shown that…” There is very little reference or discussion about the literature review. This is a dangerous practice. Literature reviews that are poorly executed or not well connected to the problem at hand can negate the relevance and importance of the study at the outset. This is a dangerous position to be in as it can set the entire study off on the wrong path. It is likened to building the foundation of a house on sand. No matter how good the study possibly turns out, if the foundation is shaky or not solid, it will not be convincing.

2. Defining the Research Question

We can now address the basic research question, the general objective of a research project, and what you are trying to do. Whether stated as a tentative hypothesis or a more open-ended question, defining it will enable you to begin structuring and planning your research. By a provisional question, we mean any question or matter on which there is doubt, or it can be a question to be solved, a theory to be tested, or a decision to be made. This can be framed as a new or refined statement of the problem and should be connected to it (see 6. Project Management) but must be based on the work that has been done in the literature review. A well-constructed question will enable you to focus your research. But how you write a question, and what type of question, will likely change during the course of the research project. This will follow from the many iterations of redefining the question and having to change tack due to what is found in the research. But it is critical to define clear and specific questions as this will affect which data is collected and will affect the design of the study. An unclear question will only lead to confusion and wasted effort. And finally, you should be thinking of some broader objectives or benefits to the scientific community or the sponsor so that you can keep in mind the relevance of what you are doing.

3. Reviewing Relevant Literature

The complex nature of the assignment necessitates an approach to knowledge production on the relevant topic area. For instance, the formal requirements of a thesis often involve the identification of specific areas of interest that are derived from a review of the literature. This process assists the candidate to understand areas that have not been explored. Furthermore, the requirement to produce knowledge is often expected of those in leadership positions. Building a robust and pertinent knowledge on the topic is crucial. Knowledge produced can inform policy and act as a basis for decision making. Undertaking a review of the literature is an unavoidable iterative process. Knowledge is never static and thesis students and researchers in general should not be afraid to expose their work to critical review. The literature review undertaken at the beginning of the candidature might be revisited at various stages in the research process, and it is expected that the student will continue to engage with the published works on their topic area throughout their candidature. So what is the literature that should be reviewed and where can relevant literature be found?

4. Analyzing and Synthesizing Findings

Analyzing quantitative and qualitative data is an important part of the review. This is a systematic and comprehensive identification and synthesis of the results of many primary studies included in the review. Descriptive and exhaustive analysis will provide answers to the questions: What is already known about the phenomenon? What are the possible inconsistencies in the results? What are the general findings? Identification of the general findings may come from an overall impression obtained while reading the study results and/or from some form of content analysis, for example, an inductive decision tree. In the latter, data from primary studies are pooled and compared with the aim of determining which variables are the best predictors of the level and dispersion parameter estimates. One way of doing this is by research on research. This can be supplemented with an analysis of the research hypothesis validity as reported at the study level. Data extraction forms need to allow data that is specific to the aims of the review to be extracted. This is an iterative process; it is rare that all of this can be done from the data presented in the published reports, and contact with study authors is often necessary.

5. Drawing Conclusions and Identifying Research Gaps

Now there are many questions that can and need to be answered. It often occurs that some literature has provided an inconclusive answer to the research question. At this point, it is important not to simply give up, as stating that no clear answer was found is still a valuable conclusion. In some cases, further research may identify a new study on the topic, and this can be the turning point. This new study may be the only addition needed to answer the question, and it is important to make an effort to revisit the literature and search for the study in due course. Answering the research question is the most important goal. A conclusive answer provides closure on the topic and serves to add to the current body of knowledge.

For many research questions, there will be conflicting or differing answers provided by several studies. These may be presented in differing contexts, and it is often difficult to identify a true answer. In medical research, this is often the case as many treatments have been compared to placebos and compared to other treatments. The need to identify the current best treatment is of top priority as this is what doctors aim to provide. In such situations, a method of critical interpretation can be used to compare differing studies. This involves a close analysis of each study in comparison to the others. You may find the need to employ further research methods, and there is the possibility of re-entering the cycle of all other aspects of a literature review. This is not a problem, and it is better to recognize this than to provide an unsatisfactory answer to the research question.

So far we have discussed convincing literature and effectively identifying and retrieving studies to answer a particular research question. The review has established the importance of this question in the current body of knowledge in terms of the strengths and shortcomings of individual studies. It has also discussed the use of various review types and the differing methodological guidelines. All of these steps are vital in answering the initial research question. If the question is not sufficiently answered, the work cannot be considered successful, and there will often be no clear answer. In some cases, it may be that the research question is not suitable and a decision to change it should be made. If the question is looked into and considered not worth asking, the literature review will have still been successful in that it led to a conclusion. The question can be shelved, and the review can provide useful analysis for a review of the question at a later date.