Too much fun with metrics? This plot shows an h-value graph based on my google scholar citations (As of December, 2017). A researcher's h-value is the number of papers written by an author that have been cited at least that many times. For me, that number is 26 (if you trust google!), which is where the red and blue lines cross in the chart. In the article below, I discuss the sorts of fun an analytical person can have with metrics. But Science is about much more than just comparing these sorts of numbers. Ultimately if you enjoy what you research, and that work is useful to others, what more do you need?
(Editorial Note: I'll be lightly editing the various posts I had created over the last year and a half as I put them up, where I've made changes, these will be noted in italics in brackets).
It's taken almost as an article of faith that a successful scientific career consists of 100 peer-reviewed publications. If that's the case, then as of November (2017) I'm halfway there Breaking the 11-month logjam at 49 papers was my methane update and, as predicted in my summer post, there has been more to come with five acceptances having followed since that time, nudging my number up to 55. If everything currently in the submitted and just about to be submitted categories follow suit, it might be 60 before the end of 2018! (indeed, it's now 62)
It seems, therefore, that this is as good a time as any for some reflection. It doesn't hurt that it is currently December (2017), the official month of reflection, and that my first sabbatical was recently approved for January of 2019. First sabbaticals are a classic break in the academic race, a time after you've established yourself to take a step back and consider where (scientifically speaking) you'd like to go next. Perhaps, then, I'm just being a bit precocious here in getting started on that longer-term thinking. Though I shouldn't get too far ahead of myself since I won't hear back with finality on Tenure and Promotion for several months yet.
Of course, metrics - despite their claim to dispassionate, quantifiable impartiality - are a fungible thing. You could build a solid argument that I'm popping the champagne on this particular celebration early. Certainly, Elsevier's Scopus and SciVal (the official research metrics of my employer, York University) would disagree with the numbers cited above. In fact, they claim my number is a "mere" 37. Meanwhile the Harvard/NASA Astrophysics Data System (ADS), a popular abstract service in planetary science credits me with 49. Google Scholar and Research Gate come up with different counts as well. Why the differences? There are primarily two factors that are important here.
The first is the Science/AAAS convention on team affiliations. A few years ago, Science decided that it had had enough of papers claiming hundreds of authors arising from what is known as "Big Science" (advances made using large international teams, typically using very complicated and expensive hardware with many governmental, academic and industrial partners). The science being presented was more than worthy of publication in such a high impact venue, but the idea that all of these people had participated in the writing of the article was obviously silly. In one famous observation, it was noted that if they had, each had contributed only one word to the main text - their names and affiliations taking up more space in the publication than did their science. The solution: restricting the author list to a smaller subset of names.
However, each of those names in the longer list had still made contributions so there needed to be a way to recognize them. As such, what Science decided was that the author list could contain a "Team Affiliation" that would be translated into the full list of names in the Supplementary Materials (a repository of data and textual description that supports the findings of a study, but which does not appear in the print version of a journal). Of my 55 publications, 13 are of this type, all associated with my work on MSL. The question of whether to "count" this information then resides with the abstract service. Google Scholar and the ADS do count these as authorship. Scopus/SciVal does not.
The second factor that explains the difference is what some might term journal "quality." I use those scare quotes not because I'm referring to predatory journals here, but to smaller publications that may not yet be indexed or that serve a relatively limited audience. For me, this mainly has to do with the InderScience journal IJSSE (International Journal of Space Science and Engineering). Despite the unfortunate-sounding name, this is a journal based out of my home department at York, where it's affectionately known as "George's Journal" after the editor-in-chief and Chair of the Mechanical Engineering Department, Prof. George Zhu. In the interests of promoting my institution and it's efforts, I have directed 4 of my papers to this venue - typically on topics which have a significant admixture of engineering. The journal is small, publishing only a few papers quarterly, and does not have an impact factor currently. That means that neither Scopus/SciVal nor the ADS index it or the publications found within. Google Scholar, however, does count it.
At the end of the day, metrics can be fun, so long as you take them with a grain of salt. I'm not going to argue that there's no way at all to compare one body of work to another body of work, but the number of different factors which need to be taken into account is significant and will vary depending on who is doing the analysis. Certainly not all papers are created equal. To give just one example, is my Nature paper a more significant contribution than my first paper in Icarus? The journal impact factor might argue that case, but others might find the latter paper (which required far more time and effort) to be more relevant to their own research. Some might suggest citation numbers to settle the debate here, but one should keep in mind that often citation rates say more about the size of the field than they do about quality. If I wanted a spectacular citation count (and hey, I'm not complaining about my 3731 [now 4869] from Google or my 983 [now 1272] from Scopus) I should have gone into medicine or particle physics.
I'm reminded this month that one shouldn't get too caught up in the numbers game. Ultimately, none of us got into science in order to out-publish our peers. Instead we can take pride in the professional, meticulous and well-considered work that we do, adding new knowledge to the storehouse and exploring ideas heretofore unconsidered.
____
As a note, I see that I originally wrote most of this post back on December 11th (2017), the day before I got sick. It has been a challenging month and things are only now starting to improve. Entertainingly, I had been telling myself that this year I would focus more on my health. It would seem that I have little choice in the matter now!
And here's another note from more than another year later, without having resolved this health issue. Sigh. Wrestling with this for a second time now, I certainly don't lack for motivation but make no predictions - look to this space in another year to find out whether I was successful.
No comments:
Post a Comment