Writing a Literature review


A literature review is the presentation, classification and evaluation of what other researchers have written on a particular subject. It is not simply a shopping list, i.e., a boring list of what others have said, however. It is organised according to your research objective, research question, and/or the problem/issue you wish to address. Without this, the literature review is a useless compilation of what other scholars have said and done. With the research objective, the literature review forms a focused and carefully structured outline of what others have done in the area that you are concerned with investigating. We cover the literature review as a writing genre here.

In postgraduate research, a literature review is a substantial work. It addresses major texts in the chosen area of study and relates and discusses their content as well as commenting on noted research or theories. The main purpose of the literature review at postgraduate level is to uncover a research gap. This is typically done in conjunction with the Introduction to the thesis. A literature review can eventually form a major chapter, or sometimes several chapters of a student’s higher degree thesis.

How a literature revew might appear in different thesis types

A literature review for a postgraduate student will consist of many article reviews combined together where the student compares article “X” with articles “Y” and “Z”, and so on, in relation to the topic he/she are investigating. They are not set out separately like an annotated bibliography, critical review, or reading log. The literature review integrates the work of others into a coherent whole. For an undergraduate student a literature review is much less extensive and demanding.

What does a Literature review do?

A literature review for a higher degree thesis summarises, interprets and evaluates existing literature in order to establish current knowledge about a subject. It often has multiple purposes. It must:

  • define and place limits on the research problem you are working on;
  • avoid unnecessary duplication and repetition by means of clustering discussion about similar authors together;
  • concentrate on key ideas relevant to the project you are pursuing, not marginal or tangential ones;
  • evaluate promising research methods;
  • relate your findings to previous knowledge and suggest further research that could be done.  

Importantly, a literature review must do more than describe. Rather, it should

  • compare and contrast different authors’ views on an issue;
  • criticise aspects of methodology;
  • note areas in which authors are in disagreement;
  • highlight exemplary studies (i.e., key works that advance the discipline);
  • highlight gaps in research;
  • show how your study relates to previous studies;
  • conclude by summarising what the literature says.  

The Literature Review may, in addition:

  • place your study in a historical perspective;
  • resolve a controversy;
  • establish the need for additional research;
  • define a topic of inquiry;
  • outline theory and list relevant hypotheses;
  • note methodological deficiencies in other studies;
  • take up recommendations for future research;
  • outline a practical problem that needs resolution.  

Owing to these quite different expectations, the literature review is a complex and difficult writing task. It takes several attempts, and multiple drafts to get it right. Below we outline a number of stages to writing a literature review.

Design a good research question

First put limits on your search of the literature by designing a narrowly-focussed research question. A good research question will not be ‘closed’ resulting in a Yes-No response (e.g., Should the constitution be changed so that the President of the US serve more than two terms? (a closed question) and admit of some subtlety: e.g., To what extent is there merit in amending the constitution to admit of changes to presidential term limits? (an open question). Seldom is anything black and white in academic scholarship. Open research questions do not admit of “yes”/“no” responses as do closed research questions—they require a discussion. This is what you want to aim for—a topic that engenders debate.

In some disciplines it is important that the research question is operational (testable); in other disciplines there is no “question” as such as as a hypothesis that must be confirmed or disconfirmed. Either way, it must be very narrowly focussed. Once you have a narrowly-focussed research question you need to consult research databases. Then you need to find a number of key papers in this narrow field of interest.

Write Annotated Bibliography entries

The next step is to write a number of annotated bibliography entries for each article you plan to review (see Annotated Bibliographies). This is essentially a summary and short commentary on each article. In doing so you should be able to see common themes in the arguments the authors provide. The next step is applying the ‘Yes-BUT’ method.

Use the 'Yes-BUT' method

This method involves setting out the literature into categories. Suppose your broad topic area is ‘Euthanasia’. You will be expected to narrow down the topic to something much more focused. This can be narrowed down to a more manageable topic such as: ‘To what extent should everyone have the right to die in a manner and time of their choosing?’). This topic requires you to review the positions for and against euthanasia. 

As you read articles on this topic you may start to agree with the proposition that people should have the right to die in a manner and time of their choosing. This becomes your tentative thesis statement. This is the position you think that you might want to defend in your literature review. But of course, you have to dispassionately consider arguments and evidence against this view. And you have to carefully weigh-up the pros and cons to this statement. In this sense, writing a literature review is similar to writing an essay.

But a literature review has to do more than argue for a position. It needs to uncover a research gap. This is the contribution that you will make in your thesis. We deal with finding the gap below. Preparatory to finding a gap, however, is canvassing the various positions in the literature that exist at present. Once this is done, it is easier to work out where new contributions are needed.

For now, write the tentative thesis statement at the top of a piece of paper (or on a Word document). Now draw a table with four columns. Title the columns: ‘YES’, ‘Yes BUT’, ‘No BUT’ and ‘NO’. These are the various responses to the assertion People should have the right to die in a manner and time of their choosing. Start reading the literature again. As you read put them into categories as follows depending on whether they agree/disagree or partly agree/partly disagree with your tentative thesis statement:

  • ‘YES’ literature agrees with the statement/question above and offers arguments and evidence to support it.
  • ‘NO’ literature does the same in opposition to the statement/question.
  • ‘Yes BUT’ literature is on the ‘YES’ side of the debate but raises major concerns or reservations about it (e.g., they may think euthanasia should be permitted by law but needs strong legislative restrictions to avoid a ‘slippery slope’ of undesired terminations).
  • ‘No BUT’ literature is on the ‘NO’ side of the debate but has majors concerns or reservations about it (e.g., they might think it should be permitted in some special cases when palliative care is not working).

This schema can be subtler than this, e.g., you might also include ‘NO but (i.e., small but)’ and ‘YES but’ literature (i.e., these are closer to the ‘NO’ and ‘YES’ positions and have fewer reservations or concerns). Set out the literature so you are clear who says what. Note the authors’ names and publication details similar to the table below.

Finding the Gap

As you read, you might start rethinking your initial response to the literature. You may even change your view entirely. This is productive as you have learned something. You might find that the issues and evidence raised are more complex and subtle than you initially thought.

Importantly, you might find that while various views are canvassed in the literature, there has been relatively little work done in certain areas of the debate. For example, you might find that while a lot of work has been done on palliative care, not much work has been done on the use of mindfulness in treating dying patients. This might influence your initial views about the topic. You might decide after a lot of reading that while in general people should have the right to die in a manner and time of their choosing, this should only be after all possible options for palliative care are exhausted. You might decide that relatively little work has explored mindfulness in treating terminally ill patients and there is good evidence for its effectiveness.

This is your research gap. You have found a unique angle on the topic. This can be where you make your contribution. Whatever position you decide to defend, ensure that the research gap is clear. This should be foregrounded in your discussion of the various points of view in your literature review.

Moving from this to actually writing the review might seem like a daunting task, but it can be made easier by producing discrete summaries, using the Yes-BUT method and following this by writing a number of critical reviews (or summary and critiques), and finally, an outline using a writing taxonomy (see below).

Write Summary and Critiques

If you are not confident moving straight to writing the literature review from the annotated bibliography entries, the next step is to write a critical review for each article. This is essentially an extension of your annotated bibliography entries, only much more detailed. You need to produce documents that both: a) describe and summarise the detailed arguments and evidence of individual published papers in your narrow area of interest; and b) critique them, i.e., note what you think is good about each article and what could be improved—or where the evidence and arguments are dubious. For this you will need to do some critical thinking. It helps to map the arguments of the various papers you are reading, and use this to map to decide which claims are objectionable.

Once you have written a critical review for each article you need to combine and assemble them in a coherent way.

Your combined document should do the following:

  • delete minor details
  • combine similar or related ideas (i.e., avoid repetition)
  • paraphrase accurately
  • reflect the different emphasis given by various authors
  • recognise the different purposes of each author
  • identify clear and common topic(s) in the discussion
  • identify a single main idea that all the authors touch on in some way that can be a source of difference between them
  • stay within appropriate length
  • exclude personal opinions

A good review will also:

  • Combine similar or related details into categories and provide a label or heading for those ideas
  • Select the main idea, or key points provided by authors when they are explicitly stated
  • Determine the main ideas when the author is not being explicit.

You want to end up with one document that combines relevant information from the relevant critical reviews already completed on each of the articles found on your topic.

It may be that the first draft of this document reads like a mess of unstructured ideas. This is normal for a first draft. It will require several drafts to get it right. The next step will help.

Design a Writing Taxonomy

The next step is to design a writing outline that gives shape to your literature review. Attempts have been made to outline writing taxonomies for graduate students that use predictable writing structures. These show nested part-whole relationships between ideas and support for ideas using commonly-used linking phrases (Rochecouste, 2005). Design taxonomies can be used to help shape a literature review. A good way to think of a design taxonomy is as a series of nested categories with ‘vertical’ and ‘horizontal’ axes. A simplified example is provided below.

A simplified writing taxonomy

This can be explained further with a more fleshed-out example. Suppose that you decide when reading the literature that there are papers that outline concrete findings (data), and papers that outline more theoretical approaches (and some papers that do both). You could cluster the findings together and cluster the theoretical approaches separately. These might comprise our vertical categories as follows:

Example of a writing taxonomy

The vertical axis in this example includes literature about the a) Findings and b) Theoretical stance on a given (unidentified) topic. These might be sub-topics of a larger issue. The nested horizontal axis outlines various positions on a) and b). They are ‘horizontal’ as they can be discussed in relationship to one another in the same, but usually separate, paragraphs. In the Findings literature, for example, they might be separated by type of finding—as they are here—or discussed in terms of chronological development (earlier to later), differences of approach, or in many other ways.

Notice the introductory comments that set up the discussion. This can be part of the introduction to your overall paper.  Or they could be separate mini-introductions to various parts of a literature review you are writing. The concluding comments could also be part of your essay or could be concluding a small section of a larger review. The writing taxonomy is applicable to a single review, or segments of a larger review.

Notice also in the example that the review begins with a sub-topic. This example therefore forms part of a literature review on a larger topic.  To take an example, the larger topic might be, for example, the role of television in influencing criminal behaviour in children. (The research question might, for example, be: To what extent does television have a negative impact on the attitudes of children towards crime and criminal behaviour?). The sub-topic might be, for example, whether violent cartoons have a direct causal impact on the anti-social behaviour in the home. The empirical work by Smith, Jones and Harrison, Davidson and Wishart, Franks and Brown is outlined and compared in the template above. A distinction is made between research findings on the impact of violent cartoons on anti-social behaviour, and the theoretical literature on the topic. Both have been given separate sections in the structure of the review. Within each section, the views of the writers are compared. A final section sums up the overall position of the writers on the sub-topic.

Naturally this is just an example, and it can be applied to any topic at all. The categories given for the vertical and horizontal axes can also be changed as appropriate to your needs. Instead of research findings and theoretical literature the vertical axes could be arguments for, arguments against, and arguments partially for/against some proposition, or they could be quantitative data, qualitative data, and mixed methods research or indeed any other way of carving up the literature. As another example, the vertical categories could be ‘theoretical approach X’, ‘theoretical approach Y’ and so on.

The vertical axes can represent the key way that scholars in the field have diced-up the literature relevant to what you are writing about. For a thesis discussing euthanasia for example, the ‘vertical’ categories might be: a) Literature supporting euthanasia, b) Literature in opposition to euthanasia, and c) Literature supportive/against euthanasia, but with reservations (this is the same as your YES-NO table outlined earlier) . The ‘horizonal’ categories would be various arguments or evidence-based considerations for each position. In the taxonomy below, the ‘horizonal’ categories are shown as stacked vertically, but conceptually they are additional examples for the same point, not new points or positions entirely. You now have the beginnings of a literature review! It is merely a matter of filling in the details.

Example of a writing taxonomy on the topic of ‘euthanasia’

Narrowing down your review

In addition to a) canvassing a variety of perspectives on the literature; b) determining where you stand on the literature; and c) identifying a research gap there should be a progressive narrowing of the literature review. The review should not get broader. It should get more refined and more focussed and detailed as you progress.

For example, if your research gap is the neglect of the importance of mindfulness as a way of providing palliative care for terminally ill patients, the review should: a) canvass the various arguments for and against euthanasia, considering and rejecting all the arguments for it, and do the same for most of the arguments against it; b) focus your attention on the palliative care argument against euthanasia and look at this in detail; c) look at the various positions on the importance of palliative care as a mitigating variable; d) focus on the various arguments for and against palliative care; e) note the neglect of mindfulness in the research literature on palliative care; f) look at the extant literature of mindfulness as a palliative care technique; g) look at what other scholars have done/failed to do; h) make your care for why it should be considered (or reconsidered). This will lead naturally to your study that is provided in the rest of your thesis.

Progressing narrowing of the literature review in two different thesis models

In this way, the literature review provides a justification for why your study is needed.

[to be continued]