I recently found a good article: A guide to writing the dissertation literature review by Justus Randolph. It’s a long article but I recommend you to read it if you are trying to write literature review too. I agree with Justus that there are not many good journal article about writing literature review, therefore I really appreciate his article even more.
There’s no denying it that writing literature review is the hardest part of the dissertation and unfortunately, it’s the most important part of the dissertation too. I once heard a speaker said that the examiners always look at the literature review first before any other parts. If the literature review is good, the examiners automatically will presume the dissertation is good. Otherwise, they will presume you are not doing the dissertation right. Scary but it’s true and justified. If the literature review is not thorough to make the argument, what can others expect of the rest of your dissertation, right? That’s why I have been quite worried about doing my literature review. I wish there are clearer steps in doing it but all I found are quite ambiguous until I found Justus’s article.
Read Justus’s article here:
Randolph, J. J. (2009). A guide to writing the dissertation literature review. Practical Assessment, Research & Evaluation, 14 (13), 1-13. Available online: http://pareonline.net/getvn.asp?v=14&n=13.
In my opinion, Justus has clearly written about the different types of review, taxonomy of literature review and most importantly the steps in doing literature review. Surprisingly, the steps are similar to doing primary research and they are as followed:
1. Problem formulation
– What questions we are trying to answer in this literature review? What criteria for inclusion/exclusion of articles?
2. Data Collection
– Justus recommends that the process of collecting articles must be documented clearly and in detail. Apparently, searches through the databases can only produce 10% of articles we want (or is relevant). The other 90% requires us to check the references of the articles (we found in that 10%) and to find/read them. (Imagine the work!) The process is not completed until we reach saturation (which means no more new articles found during this cross-reference).
3. Data Evaluation
– Once we have collected ‘enough’ articles, we have to extract/evaluate information in the articles based on the inclusion criteria that we set earlier. Justus advises to use a ‘coding book’ to record all types of data to be extracted. You have to read his article to find out more because I’m not expert in this yet. However, I believe his idea of coding book is similar to literature matrix that my faculty mentioned.
4. Data Analysis and Interpretation
– With all the data extracted, it’s time to make sense of everything.
5. Public Presentation
– This part is where we start writing, I guess. But first we must find out what’s important to be put in writing and present to the world.
Simple and clear steps but I guess can take very long time to complete because it depends how big your literature review is. And for me, after reading this, I’m no way closer to completion of literature review any time soon. Sigh!
In order to synthesize your sources, you must first analyze them to help provide rationale for why they are a part of your literature review and what role they play within your field. It can be difficult, however, to demonstrate analysis without inserting one's opinions or beliefs.
Consider the following, analysis-free excerpt. This approach is typical of a student looking to avoid opinion in the paper:
....is to deny the student (Sigree, 1999).
As Harper (2001) noted, instructors cannot identify every one of their students' emotional intelligences (EI). Faculty members do not have the time, and students simply are not that forthcoming with their learning preferences (Harper, 2001). Furthermore, as Harper warned, if instructors decide to attempt a complete analysis of every student's EI, they will inevitably hold the entire class back. After all, taking time to adequately diagnose a student's EI means less time for helping students meet the expectations set forth by the No Child Left Behind Act (Harper, 2001).
Finkelstein and Kramer's (2002) findings…
There is no analysis or critique in this excerpt. There is strong paraphrasing, and this passage provides a decent overview of Harper, but it addresses only Harper's ideas and does not explain why this information is important and how it relates to the author's overall purpose for the paper. The reader needs to know the answer to "So what, and who cares?"
What is missing from this summary is context and analysis. Consider the following revision:
....is to deny the student (Sigree, 1999).
Harper (2001), however, disagreed with Sigree's (1999) assertion. Harper noted that despite the obvious benefits of diagnosing a student's emotional intelligence (EI; Jones & Hammer, 1998; Mooney, 1998; Sigree, 1999), instructors cannot identify every one of their students' EIs (Harper, 2001). Faculty members do not have the time, and students are not that forthcoming with their learning preferences (Harper, 2001).
For Harper (2001), though, the real issue was not with instructors' belief in EI, but rather in how this belief affected classroom logistics. Instructors who follow Earnhart's (1996) advice to "Take the time to understand how each of your students learn" (p. 33) are being impractical, Harper argued. Taking time to adequately diagnose a student's EI means less time for helping students meet the expectations set forth by the No Child Left Behind Act (Harper, 2001). Although Earnhart's (1996) vision is ideal, Harper takes a more practical stance.
With Harper's (2001) concerns in mind, I cannot endorse Finkelstein and Kramer’s (2002) findings…
Here, the author is synthesizing the literature. We know, based on the author's direction, how Harper interacts with the other literature on the topic. We know that Harper is probably in the minority, and we know what the author's take on Harper is. Finally, we know how and why the author is using Harper: Harper will be used to refute Finkelstein and Kramer, which presumably is the author's intent or thesis. Notice how the author demonstrated analysis and synthesis in just a few additional sentences. Note, also, that the author includes an "I" statement in the passage but did not insert his or her opinion.