How do I go about writing a review?

Question

I'm an agronomy major who's performing additional academic work as way of increasing my chances of getting into grad school.

This semester I'm going to be writing a literature review regarding plant breeding in ornamental plants using references from 2007 to the present year. The emphasis will be in breeding methodologies ((1)mutagenesis, (2) genetic transformation) and (3) biodiversity.

My professor has done extensive research in genetics but this is also the first time he's advising a student in writing a review. Out goal is to make it as publishable as possible.

I've already obtained multiple articles but this is my first time writing a review. Can I get advice on how to start or how to plan my creation process?

Answer 1

First, you'll need to identify what kind of reviews you're planning to write. Just to name a few: critical review, integrative review, literature review, meta-analysis, scoping review, systematic review, etc. The methods differ somewhat regarding the types. For example, if you're writing a systematic review, then you'll need a much heavy focus on how you searched for the articles and how you sorted/selected them.

Second, most reviews are question-driven and according to the types, the specificity of the questions also differ. Scoping reviews may have a looser definition of questions while meta-analysis and systematic review may have a set of tightly defined questions.

In other words, it's likely not something we can introduce on a Q&A website like here. As a "review" varies so much that can be as quick as an afternoon of search or a five-year PhD project.

What I'd recommend:

1. Identify about 5-8 review articles in your field that you'd like to model after. Most reviews are pretty apparent in terms of what type of reviews it is. But if you're unsure, talk to a librarian.

2. Read up about that kind of review. There are books (like Booth's Systematic Approaches to a Successful Literature Review) and guidelines (check www.equator-network.org) on how to carry out the search, arrange ideas, and present your interpretation. Again, your library would be a great start.

3. Consider working with someone who had experience in writing a review. They can help you with revising subtle things like how to tabulate the data to make them digestible and useful, and how to present your findings, etc.

4. Budget a lot of time.

5. Develop a system of keeping good notes and bibliography. People have used index cards, highlighters, Evernote, etc. to keep their thoughts and their quotes clearly laid out and discernible from each other. Bibliography management software is also a must, as there will be intensive referencing.

6. If you have tuition remission/support, consider register for a course in your school on literature review. Most of these courses require students to have an independent project (you may petition for one if there isn't.) That way you can get to work on your topic and obtain guidance and input from the course instructor. I think 13 to 14 weeks of weekly effort should get you a very strong draft. (According to your school policy you may want to give the instructor a heads up and seek approval on using your research topic, though I don't see any conflict here.)

What I'd privately recommend:

1. Evaluate if you truly need a review. While I understand it sounds logical to "read up and write about the field" before you joint in. It's true that it's easy to read a lot of articles about a field, yet it takes time/experience to be able to tell the intricacy among them. So, most reviews take a much, much longer time to write compared to, say, writing up about an experiment. If you don't have much time before the application, consider at least another shorter paper such as a part of your PI's research that you are involved in as a side project.

2. I am not at all familiar with agronomy, but I'd guess that the topic may be already too broad for some kind of reviews as it covers all "ornamental plants." Dissect the sample articles you got and examine how the authors specifically model their question statements.

3. Most common newbie mistake is writing an annotated bibliography. If the review reads like "A says a, B says b, and C et. al. says c," chance is it's not being done quite right.

Answer 2

Please excuse me for recommending my own work, but I've done quite a bit of work to address the problem: "I need to do a literature review. How do I do it? And once I've collected a bunch of articles, what do I do with them?"

In brief, I recommend adopting a theory-mining approach. That means to explicitly look for theory in the source articles and then to synthesize the theory as the contribution of the review. What I mean by "theory" is very simply "A is related to B". This is distinct from literature reviews that simply summarize what people have said about A; that is not theory. Theory is about explaining how and why phenomenon A is related to phenomenon B. In my academic field (information systems), I have found that this focus on theory is the most useful thing that can be done with literature reviews (in terms of citations by other researchers). So, in your case, for example, you might explicitly focus on literature that studies which breeding methodologies effectively result in certain desirable characteristics. Although my literature review approach focuses on social sciences, I believe it would be relevant in a case like yours.

I have two main relevant articles to recommend:

• The View from Giants’ Shoulders: Developing Theory with Theory-Mining Systematic Literature Reviews: this describes the idea of theory-mining reviews and focuses mainly on theory
• A Guide to Conducting a Standalone Systematic Literature Review: this provides more practical details on how to actually conduct a general literature review