Engineering Management

What do your professors want in a literature review?

Whether you are doing a topic summary for a term paper, a state-of-the-art survey, or a full literature review for a thesis or article, there are some common expectations that your professors have for graduate student work. They are not looking for you to simply describe some papers that you have read on the topic, one after the other. What they do expect is:

  1. That you have found and thoroughly read enough papers to have a solid grasp of the particular topic. This is where it's very important to properly define your topic so you can do a good job, and do a structured database search! You should start to encounter some of the same authors and papers repeatedly as you read, indicating that you are finding the major works in this topic. For searching advice, see the Find Articles tab. You should use at least two search tools (Scopus, Web of Science, Google Scholar, etc).
  2. That you have understood them enough to identify major trends, methods, approaches, and differences. This takes work! You do not want to just re-phrase the abstract. See below for some tips on doing this.
  3. That you can communicate your own perspective and informed opinion on what is truly important - including where the current research is lacking (where there is a gap). If you are doing your own research, this is a very important part of the literature review as it justifies the rest of your project.

The process of doing a literature review

Process of doing a literature review

Source: North Carolina State University. (n.d.). Literature Reviews: An Overview for Graduate Students. https://www.lib.ncsu.edu/tutorials/litreview/

Reading and note-taking efficiently

Getting started

You want to be organized from the start when doing a literature review, especially for a project that will take a long time. 

  1. In a Word or Excel file, keep track of your searching - which search databases and tools you use, and paste in all the search queries you run that are useful, with parameters. In Scopus, for example, this might be 'TITLE-ABS-KEY ( anaerobic  AND digestion  AND feedstock )  AND  PUBYEAR  >  2013'. This will help you avoid duplicating work later.
  2. Use a citation manager program like Zotero or Mendeley, to keep track of your papers as you find them, and format citations later. See this guide for details on the programs. Save the PDFs to your computer, and attach them to the entries in your citation manager if it isn't added automatically.

Reading and Note-taking on Individual papers

When you actually read the papers that you find, most people take a staged approach to save time:

  1. Read the abstract fully to determine if it's actually on topic.
  2. If so, read the discussion and conclusion, and the figures and graphs, to figure out if the results were significant or produced interesting results.
  3. If so, make sure it is saved. Then read the full article, and annotate the article right away.

What does annotating mean? Take very short notes (on paper or digital) of the most important findings and/or highlight important lines in the paper. You can highlight and annotate the PDF file if you want, or in your citation manager. You don't usually need to summarize the whole article - instead focus on what is important for your research or review, and write it in your own words. This could be the

  • whether the study was theoretical, experimental, numerical simulation, etc
  • main theoretical approach, model, algorithms, etc
  • number of test specimens or subjects
  • key assumptions made that might impact its general validity
  • key outcome measured, statistical significance of it, etc
  • Your own comments - for example, strengths and weaknesses

Synthesizing the papers and structuring your review

Concept mapping

One technique is to create a concept map or 'mind map' showing the relationships or groupings of the key papers on your topic, with short labels. This way, you can try out different options for how to structure your paper and see which one makes the most sense. You can do this on paper:

You can also do this digitally, using a mind-mapping website. There are some easy-to-use, free tools that are available now. Two that I have used are Coggle and Miro. You can also just sketch on paper.

Mind map showing papers for the topic 'methods for bearing signature extraction'

Created using  Coggle.it, based on a chart in Huang, H. (2018). Methods for Rolling Element Bearing Fault Diagnosis under Constant and Time-varying Rotational Speed Conditions (Ph.D. Thesis, University of Ottawa). http://dx.doi.org/10.20381/ruor-21835

Image: Pacheco-Vega, R. (2016, June 15). How to do a literature review: Citation tracing, concept saturation and results’ mind-mapping. Retrieved from http://www.raulpacheco.org/2016/06/how-to-do-a-literature-review-citation-tracing-concept-saturation-and-results-mind-mapping/

After you have taken notes on individual articles, it can be very helpful to create a chart with key variables that seem important. Not every article will cover the same material. But there should be some common factors, and some differences between them. This chart is called a synthesis matrix.

Example of a 'synthesis matrix'

  Chen, 2012 Brooking, 2015 Sharma, 2016
Preferred materials

cadmium telluride (p. 312)

copper-indium selenide (p. 1209)

polycrystalline silicon ( 54)

Efficiency of solar cell

12% under STP (p. 65)

15% (p. 1215)

22% at 45 deg. C ( 56)

Use of thin-film solar cells

depending on application, can be preferred (p. 320)

cannot be used above 50 degrees Celsius (p. 1213)

not preferred - cost to efficiency of silicon is higher (p.  59)

Source: University of Western Ontario Library (n.d.). “Writing your literature review”. https://guides.lib.uwo.ca/mme9642/litreview

See this blog post by researcher Raul Pacheco-Vega for another example of how he does this.

This chart can help you decide how to organize your review. If it's a very short review, some people write it chronologically - they describe how the topic evolved, one paper at a time. But if you have more than 10 papers, this is not a good approach. Instead, it is best to organize your review thematically. In this approach, you group the papers into several groups or themes, and discuss each theme in a separate section. Usually the groups are major methods of tackling the problem, or concepts, or techniques.

In each section of your paper, you introduce the theme, and then discuss and compare the papers in the group. Using this approach lets you show that you have not just read the papers, but have understood the topic as a whole, and can synthesize the literature.

For example, this paper co-authored by Ping Li, a Civil Engineering PhD graduate of uOttawa, organizes the papers into three categories: ones that used a 'traditional' approach; ones based on characterization of the soil microstructure, and ones that also incorporate soil mechanics. The strengths and weaknesses of category are discussed, and in the conclusion, the authors recommend approaches for future studies. 

You can often include a form of a synthesis chart in your paper or thesis, as a visual summary of your lit review. This is part of a chart included in a Masters' thesis in Computer Science from uOttawa.

Part of a chart showing various papers on Phishing Detection.

From Le Page, S. (2019). Understanding the Phishing Ecosystem (M.Sc. Thesis, University of Ottawa). http://dx.doi.org/10.20381/ruor-23629