3

In conducting a literature review i.e., research done on a particular topic for a certain period of time what is the prefered period of the review? Would it be advised if the time period is from 2000 to present only?

  • 4
    The time since the last review paper. :-) – Matthias Feb 6 '18 at 15:36
  • But there's no review paper done before. – xavier Feb 6 '18 at 15:42
  • 6
    Then it is a bad idea to limit the period strictly by time. You should start with the research that defined the area, and follow at least the main lines since then. – Matthias Feb 6 '18 at 15:46
14

The year that a paper was published is entirely irrelevant to the question of if it is worthy of inclusion. I’ve read good review papers citing work that is almost three thousand years old and I’ve read good review papers not citing anything older than 2001. Using age as a proxy for relevance is an incredibly poor decision. You should include what is important for someone in the field to know, regardless of when it was published.

3

The simple answer is that you should include all relevant research, regardless of date. However, there is a theoretical aspect to this, which is contrasted with a practical aspect.

In theory, or in principle, a literature review should go as far back in time as there are still relevant articles. On one hand, if there are older articles that are superseded by newer articles that incorporate their older information along with new updated information, such older articles need not be cited, since that would be redundant. On the other hand, it is a general academic principle of courtesy to always cite the original source of an idea (if it can be clearly established who originated the idea), in order to give credit to the original source--this might involve citing a few very old articles.

In practice, however, the practical execution of all literature reviews are constrained by the amount of time that the reviewer has to do the work and by the amount of financial resource available to do it (e.g. whether hiring assistants is an option). If the constraints only permit reading and doing a proper review of only 50 articles, then it might be more efficient to set a cutoff date so that only the most recent articles are reviewed. However, setting a cutoff date requires the reviewer to prove that there is a logically sound reason to expect that older articles would not add relevant information. (For example, it could be reasonable to limit a study of online education to articles older than 1989 on the grounds that only Web-enabled online courses are of interest, and the Web was invented in 1989.) Without such an objectively clear justification, then the review is probably incomplete for no acceptable reason.

That said, even if there are such practical constraints, I would strongly advie the reviewer to include a few older articles that are highly cited by more recent ones, especially if such older articles seem to have important information for understanding the evolution of thought in the area. Above all, there should be no fear that if the reviewer sets a rule for articles 2000 and after, then it is a "sin" to include a few articles older than that. It is OK to set practical boundaries for your search criteria, but to exceed your boundaries for specific exceptions that pop out to you. That is good literature search practice.

  • older articles need not be cited, since that would be redundant — Except to give credit to the original work, of course. – JeffE Feb 7 '18 at 14:23
  • @JeffE That's exactly what I say in the next sentence. – Tripartio Feb 7 '18 at 17:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.