By Julia Hallas, PhD Candidate
“Never use references that are older than three years” was the advice given by a journal article reviewer I went along to hear recently. Yet a postgraduate supervisor who was sitting next to me whispered, “except for seminal papers”. I have heard a number of conflicting opinions about date ranges and types of references to use in a thesis or journal article. Yet such advice can limit one’s thinking and creative capacity, so I think that it’s worthwhile thinking through a strategy and a rationale for the references you decide to include in your research project. Let’s think about how we might define papers as seminal, and if we should ever use references that are older than three years.
Cite influential theories
Seminal papers are those that explain well-known theoretical frameworks and concepts. Such a theory will have been influential in developing a field of study and will be well cited in the literature. For example, I am drawing on Vygotsky’s (1978) ‘zone of proximal development’ theory in my thesis. Many researchers have studied this significant concept over past decades, and I could cite one of those papers, however it would be as a secondary source. Much better for me to build my academic knowledge and integrity by reading Vygotsky and citing him as the original source.
Cite the latest empirical studies
What kind of date range should we be looking at when searching for empirical studies? Do you know how literature is selected in your discipline? I looked for advice in three useful texts on research. In describing how to undertake a literature search, Creswell (2009) did not mention a date range at all; while Merriam (2008) and Cohen, Manion & Morrison (2011) both suggested that only the most recent work in the area should be included in a literature review. There are so many journals and publications today, it would be very difficult for anyone to read all the research on a subject. Therefore common sense must prevail. A sensible strategy would be to discuss with your supervisors a date range that would provide the most up to date work for your study. However once the date range is determined, there is no need to stick rigidly to this. For example, if you notice an older reference that is often cited often in the articles you are reading, check it out. If you still think it’s worth including, cite it as a primary source.
So there you have it. Older articles have their place in thesis writing, as do the most up-to-date empirical studies. Think about the approach you might take and discuss it with your supervisors. They will be pleased to see that you are thinking about this important element of research: how to deal with older and newer references in your thesis writing.
References
Cohen, L., Manion, L., & Morrison, K. (2011). Research methods in education (7th ed.). Oxon, UK: Routledge.
Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). California, CA: Sage.
Merriam, S. B. (2009). Qualitative Research: A guide to design and implementation. San Francisco, CA: Jossey-Bass.
Well Julia thank you for this post. I am finding that many of my references are old indeed especially when dealing with fundamental philosophical issues as many of today’s dilemmas have been explored before. If one accepts a cyclical construct of time and thus Heraclitus, Parmenides and Plato from over 2000 years ago are as relevant as the Upanishads. Ah, but the more recent works of Leibniz, Hegel, Husserl and Sartre are as relevant as Badiou, Latour and Zizek!
I enjoy reading Julia’s posts – always useful topics. Consideration of the age of references is important, but too much emphasis on the age of the reference can overshadow a more important criterion – relevance to the research question. I usually gather what seems relevant – and then look at the time frame covered. That can indicate some gaps but can also tell a story in itself.
I could add two additional things:
• Does the issue of age of references vary between disciplines? Are there fast moving disciplines such as IT, where older references may have less relevance?
• The history of thought can be interesting and/or important. Maybe some of the older thinking may not make it into your thesis, but could be important in defending why a particular approach has been taken. The development of ideas over time on a topic can be an important finding.
I have a question – at what point do you stop? How close to the submission of your thesis can our literature search be expected to cover?
Imagine a scenario – you are a month or a week from sending your final version to a proof-reader. You and your supervisor are happy with what you think is a final version of your thesis. Another article comes out from a highly respected author of considerable standing in the field. What do you do?
Thank you Ross and Katharine for your comments.
Katharine I hope a supervisor answers your question because I would like an answer to that one as well. Cheers, Julia
In your thesis write the last date that your literature review was updated. Don’t re-write your thesis if a new publication comes out after the cut-off date unless the new publication is so fundamentally contrary to what you have written in your thesis that it needs an answer. This is almost never. If it is an important update then as a researcher with an ongoing interest in the field then you will naturally have read the publication and have an opinion on it that can be stated when asked.
Well this depends. Older references can be outdated. Still, older sources may be more relevant than newer ones. I guess it all depends on the topic. Basically, there is no formula.