Doing literature surveys: pointers to some resources

Many students and researchers will do a literature survey as part of their research project. Literature surveys are essential steps towards forming a good research question and clarifying your research contribution. In this page you will find some articles that I recommend you to read in order to turn your literature survey into a piece of research contribution, and get an A+ grade in the process! This page is work in progress. More references to be added.

Writing a literature review: Introduction

If you have no idea what a literature review is probably you can start reading this excellent introduction from univ. north Carolina at Chapel Hill. There is also an excellent introduction to different types of literature reviews (narrative and systematic) written by  Patrick Dunleavy at this page. As a more advanced introduction to the field take a look at Webster and Watson’s paper from MISQ 2002. Pay special attention to their recommendation to organize your literature review not as a summary of the papers you read but as a map of the research concepts that are discussed in those papers. An interesting paper that discusses a method for classifying and presenting findings and topics from literature using qualitative methods and grounded theory is Wolfswinkel, J. F., Furtmueller, E., & Wilderom, C. P. M. (2011).

If you are interested in knowing what is the value of literature studies and how they are evaluated for quality take a look at Boote and Beile (2005).

Narrative literature reviews

Narrative literature reviews are those that provide a more or less organized story about what has been done earlier. They allow some degree of creativity with respect to the organization of the story. Narrative reviews can have a chronological, topic-based, dramatic or other type of flow. Narrative literature reviews are the most common form of reviews. Most students write a more or less extended narrative review as the “related work” part of their thesis. Narrative reviews are more subjective and depend a lot on the understanding and the objectivity of the writer. Narrative reviews have been criticize for being biased. Authors sometimes cherry-pick literature that supports their view and leave out literature reporting opposing views. The article by Patrick Dunleavy at this page gives some more clues about this type of review. Webster and Watson’s paper from MISQ 2002 is also applicable to narrative reviews. Also take a look at this article on realist reviews, which is a method for qualitative reviews in the borderline between narrative and systematic reviews.

Systematic literature reviews

If you take your literature review seriously you can go for a systematic literature review. This is a form of a survey study where you formalize the search and inclusion processes and use rigorous coding and classification. The de facto guide for doing this type of reviews in ICT is Kitchenham, B., & Charters, S. (2007). For an example of a well-known systematic literature review from the software engineering field see Dybå, T., & Dingsøyr, T. (2008).

Systematic mapping studies

Mapping studies are a different type of systematic surveys that look into a larger field of research and map the research done in such a field. The difference with systematic literature reviews is that mapping studies are broad while reviews go into depth. Mapping studies can be quite large. There are a couple of guides available. Clapton, J., Rutter, D., & Sharif, N. (2009) is one non-ICT guide with a lot of detailed instructions. Petersen, K., Feldt, R., Mujtaba, S., & Mattsson, M. (2008) is a shorter guide with a focus on software engineering and information systems. A recent example of an ICT-related mapping study is Farshchian, B. A., & Dahl, Y. (2015), which also has a discussion of the differences between mapping studies and systematic reviews.

Using search engines

When you do literature reviews you normally have to search online for literature. Google is not a good choice because it is not specialized in finding research literature. Google scholar is a better choice. See also the article by Patrick Dunleavy that has a lot of good tips on how to start searching using Google Scholar. Especially the citation number field shown in Google Scholar for each resulting paper is a very useful clue about how important the paper is: the higher the citation number divided by the age of the paper, the more important the paper is. If you use systematic reviews and mapping studies you might want to look into a couple of papers that discuss the usefulness of search engines: Dieste, O., Grimán, A., & Juristo, N. (2009) and Bailey, J., Zhang, C., Budgen, D., Turner, M., & Charters, S. (2007).

When you do search, you need to create a search string. A good search string is essential, and is closely related to the research question that you have defined for your project. A paper that discusses the issues related to creating a proper search string is Cooke, A., Smith, D., & Booth, A. (2012).

Link to other relevant resources

  • If you are writing a systematic review, you might consider using this template that has all the relevant sections often required in such reviews.

Miscellaneous tips

  • When coding papers, use the framework by March and Smith (1995) to categorize type of contribution and type of research activity.
  • Do some pilot search and coding before you finalize the search. You will then learn a lot about the problem and create better search strings.
  • I spend approx. a whole day for coding 100 abstracts. You should know your estimate so you don’t get surprised. Keep time when you do coding so you can estimate how log it will take.

 

Leave a Reply

Your email address will not be published. Required fields are marked *