How to...
write a literature review
To write a literature review it is important to look at the relationships between different views, draw out the key themes and structure appropriately. See our step by step guide for some useful tips.
What is a literature review?
Essentially, it is a description of work that has already been published in a particular field or on a specific topic. There are two main types of literature review:
Research literature review – This doesn’t contain new research but looks at experiments already published and reports on their findings. It gives an overview of what has been said, who the key writers are, the prevailing theories and hypotheses, the questions being asked, and the methods and methodologies that have proved useful.
Systematic or evidence-based literature review – Especially popular in medicine, these reviews are designed to find the best form of intervention, or explore summaries and critiques leading to better future practice.
Why write a literature review?
For students, a literature review is often part of a thesis or dissertation, forming an early context-setting chapter. For academics, it is a necessary part of a research paper, setting the scene and showing how their own work contributes to the body of knowledge.
This guide focuses on literature reviews that go on to be published as individual journal papers.
Research literature reviews
The format can be purely descriptive, i.e. an annotated bibliography, or it might provide a critical assessment of the literature in a particular field, stating where the weaknesses and gaps are, contrasting the views of particular authors, or raising questions.
Whichever format you choose, it’s crucial that the review doesn’t just list and paraphrase the content of the papers involved – it should also show evidence of evaluation, and explore relationships between the material so that key themes emerge.
Systematic or evidence-based literature reviews
These use explicit and transparent methods. They follow a standard protocol, or series of steps, often established in consultation with a panel.
All procedures are documented, i.e. there is a research audit trail of databases and search terms used, so that others can easily replicate the steps followed. The documented procedures might include:
- The search parameters
- Databases used
- How papers were analysed
- Criteria for inclusion in the final review
The systematic review was originally developed in the field of medicine, through the Cochrane Collaboration (Hemsley-Brown and Oplatka, 2006). While its origins lie in the field of evidence-based healthcare, it has also been adopted by some researchers. Because of its rigorous approach and transparent methodology, it helps to eliminate bias from the selection of literature.
Creating a literature review - step by step guide
The following steps apply for all types of literature review.
It’s important to establish a purpose for your literature review but the key is in finding the right balance – too narrow and you will have limited sources to review, too broad and the list will be endless. Some authors choose to confine their review to a specific time period.
It's important to be systematic - whether you follow a list of database references, or jump directly to the citations of a particular article, you need to keep records. These should not only be bibliographic (author, date, title of article/chapter, publication, volume and issue number, edition, etc.), but focus on the content too.
Searching
Keywords are a good search strategy and it helps if you are specific (don’t rely on general keywords and phrases). You can also search for key scholars in the field by name.
Too many results returned? Try filtering using the following methods:
- Metadata: Refine material by author, year of publication or geographic location.
- Semantic: Remove words or terms that are spelled the same but differ in meaning.
- Evidence-grading: Apply a quality filter e.g. sift out non-peer reviewed, or opinion-based, rhetorical, and non-conclusive material.
- Accessibility: Is the full-text article available or just the abstract?
Sources
You will want to search for relevant materials across a range of media. Possible sources include:
- Books (monographs, text books, reference books)
- Journal articles
- Newspaper articles
- Historical records
- Commercial reports and statistical information
- Government reports and statistical information
- Theses and dissertations
If you are attached to a university, the library is probably the best place to start. You could also refer to other relevant library catalogues, such as the British Library catalogue, the National Union Catalogue (Library of Congress), and, through their URLs, other large academic libraries. Most libraries will also have indexes of periodicals, e.g. Business Periodicals Index, and abstracting services, e.g. Dissertation Abstracts.
While there are special circumstances for using old sources, for example in a historical study, or because the work is seminal, ideally you want to focus on the most recently published literature.
When you are looking at your raw bibliographical data, there are some important points to consider:
- What are the author's credentials? Are they an expert in the field? Are they affiliated to a reputable organisation?
- What is the date of publication? Is it sufficiently current or will knowledge have moved on?
- If it’s a book, are you looking at the latest edition?
- Is the publisher a reputable, scholarly publisher?
- If it is a journal, has the content been peer reviewed?
As you move on to analysing the content, your questions change in tone.
- Is the writer addressing a scholarly audience?
- Does the author review the relevant literature?
- Does the author write from an objective viewpoint, and are their views based on facts rather than opinions?
- If the author uses research, is the design sound?
- Is it primary or secondary material?
- What is the relationship of this work to other material on the same topic? Does it substantiate it or add a different perspective?
- Is the author's argument logically organised and clear to follow?
- If the author is writing from a practice-based perspective, what are the implications for practice?
- What themes emerge and what conclusions can be drawn?
- Are there any significant questions forming a basis for further investigation?
The Cornell University Library website contains some good pointers for evaluating material, including how to distinguish scholarly and non-scholarly publications.
Between the first and second stages, there should be a process of selection; not everything you read will go into your final literature review.
One useful way to find common strands and show up apparent contradictions is to create a table of your results with study references listed alongside a brief overview of findings. These could be:
- Statistical – results subjected to a set of statistical tests, i.e. meta-analysis.
- Narrative – organised by theme, study type, etc.
- Conceptual – different concepts brought together and a new concept described.
There are many ways to organise a literature review. Let’s take a closer look at one option:
Introduction: Define the topic, together with your reason for selecting the topic. You could also point out overall trends, gaps, particular themes that emerge, etc.
Body: this is where you discuss your sources. Here are some ways in which you could organise your discussion:
- Chronologically: For example, if writers' views have tended to change over time. There is little point in doing the review by order of publication unless this shows a clear trend.
- Thematically: Identify a series of themes.
- Research type: For example, academic versus practitioner.
- Dialectical: Contrast different views or theoretical debates.
- Methodologically: Here, the focus is on the methods of the researcher, for example, qualitative versus quantitative approaches.
As with any piece of writing, make sure that your structure is clear by explaining what you are going to do, and using appropriate headings.
Conclusion: Summarise the major contributions, evaluate the current position, and point out flaws in methodology, gaps in the research, contradictions, and areas for further study.
