Sitting in a meeting and listening to a paper on a subject I know fairly well, I became suddenly aware that a lot of the material that was being presented I had heard before.
I nudged an old colleague who was sitting beside me, and asked, "Didn't so-and-so do something like this about 25 years ago?"
He nodded and said yes, that it was right after the war--World War II--when he was at the U.S. Air Force labs in San Antonio. Knowing that my colleague was right, after the presentation, I asked the speaker for a copy of the paper. In perusing the references, I noticed none from the post-WWII research, and, moreover, the earliest reference was 15 years ago. He didn't know of the earlier work.
This is not an isolated incident. In fact "having heard that song before" seems to be occurring with greater and greater frequency.
How does this happen? Suppose I'm delving into an area of research--one that was fairly new to me. The first thing I'd do would be to check the references in several of the standard literature search files, giving my searcher several key words to home in on. The result would be a listing of all the relevant articles. But the factor that is critical in this whole procedure is the question my searcher asks me--"How many years do you want to go back?--10 years, 20 years?" And so, like the speaker I had heard, I'd reply, "Oh, 10 years is good enough."
And this is why I had heard about the research "before." It was conducted more than 20 years ago.
I review papers for three journals and the majority of the manuscripts are excellent and require little editing. But occasionally, and not too frequently, I run across reports on research that, if not conducted earlier, certainly have employed similar methods and obtained results similar to those of previous studies. I have brought this observation to some of my senior colleagues and they, too, have had similar experiences.
So, where does this get us? To my younger research colleagues, may I make this suggestion? First, before launching a research study, see if it or something similar has already been done. A good place to start is the Annual Reviews. If you uncover similar studies, get the original article and examine the procedures, techniques, suggested modifications and interpretations of the earlier work.
Then if your proposal is unique and the statistical treatment of the data appears appropriate and valid, go for it. And, when your proposed study is ready to be launched, get in touch with one or two emeritus professors to look it over and make suggestions. Their reinforcement--whether positive or negative--can only enhance your effort.
For example, the human factors and ergonomics graduate students here at Kansas State have a "brown bag" colloquium every Monday at noon. Fortunately for me, I have been included and I have enjoyed them immensely; they keep me abreast of some of the latest thinking in the field, which to me has been both stimulating and exciting.
However, on several occasions I have suggested minor modifications in the experimental design or the statistical treatment of the data that may have been overlooked. Remember "what goes around comes around"--a point that is paramount to avoiding the pitfalls of duplication.