Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456

Address: [go: Go Back, main page]
Remove Scripts Accept Cookies Show Images Show Referer Rotate13 Base64 Strip Meta Strip Title Session Cookies

How Should One Evaluate Collaborative Research?

16 September 2004

According to both the DBLP Bibliography Server and MathSci, I am "over fifty". A look at my publication list indicates that most of my research output consists of papers based on collaborative research, and I have already expressed some of my opinions on, and fondness for, research collaborations in my first posting. To sum it all up, I firmly believe that, even when, given time and stamina, I could have possibly written a paper alone, the joint paper that resulted from the collaboration with my co-authors was substantially better than what I could have possibly achieved alone. It is a truism, but many brains and pairs of eyes are better than one.

Collaboration in research is increasing all the time. Indeed, the study of the structure of scientific collaboration networks is becoming a very active field of study. (See, e.g., the paper The structure of scientific collaboration networks (January 16, 2001 edition of the Proceedings of the National Academy of Sciences) by Mark Newman, and the work on Erdös numbers.) What concerns me here, however, is not so much the study of these networks, but rather the following question:

How does one evaluate collaborative papers?

In particular, many of us sit, at some point in their career, in evaluation committees whose aim is to choose the next (associate) professor to be appointed at some institution. If an applicant comes along, and we like his/her publication record, is it meaningful for us to cast doubts about the research qualities and stature of this person just because (s)he has a good network of prolific collaborators with whom (s)he has co-authored many papers? If a researcher has written papers with several colleagues who are more prolific than her/him, does this mean that (s)he did not contribute significantly to the research?

Several institutes require co-author statements to accompany the submission of a doctoral dissertation that contains papers that the PhD candidate has co-authored with other researchers. (Even though common wisdom would suggest that it is the PhD student who has carried out a lot of the work on the paper --- if only because (s)he has more time on her/his hands than the senior researchers, who usually have a lot on their plate.) If, for some reason, an evaluation committee has any doubts about the actual contribution of the researcher under evaluation to the papers (s)he co-authors, then such practice could be extended to evaluations for senior academic positions. Letters of references also serve a useful purpose here.

I must be naive, but I believe that being a member of an active and prolific network of collaborators, with an ingrained habit of writing papers together, is a very positive thing! To my mind, this is the best insurance against research burn-out, and prolonged droughts in productivity. (As a colleague of mine once said during our last research evaluation "Lone wolves die!")

For this reason, and possibly my naive attitude, when I see a paper co-authored by several researchers, I do not ask myself "Who was that actually did the work?", but rather end up thinking that "They are doing well! Here is another paper to read."

Another lesson that I hope to learn, and put into practice during my evaluation jobs, is that, when evaluating the work of our colleagues, we should realize that we are commenting on somebody's life work, and thus professionalism and the respect due to a learned peer are in place. Let's bear this in mind, when we pass judgment on other people's output, and on their collaboration graph.

Let's collaborate and write papers together! Contact me if you wish to write a paper with me, and get a finite Erdös number to boot!