The 2009 Race for Impact by Journals in Sport and Exercise Science and Medicine, and Tom Reilly's H Index
Sportscience 12, 24-27, 2009 (sportsci.org/2009/wghif.htm)
Updates July 2009: Hirsch (2005) demonstrated that his index is approximately proportional to a researcher's number of productive years. Researchers in the same discipline can therefore compare their H indices by dividing them by their years of publishing. Tom Reilly authored his first paper in 1973, giving him an H "rate" of 38/36 = 1.06 per publishing year. Several minor errors in the section on the H index have also been corrected.
This article represents my annual update of the impact factors of journals in the disciplines of exercise and sport science and medicine, cribbed from the latest edition of Journal Citation Reports. If you are new to the notion of a journal's impact factor, it is the average number of times the average article in the journal has been cited recently. As such, it is an objective measure of the credibility or usefulness of articles in the journal. The impact factor probably adds little to what experienced researchers already know about the relative importance of journals. Nevertheless, these updates are justifiable if only for their entertainment value, which is not dissimilar to that of competitive sport. See last year's update and the links therefrom for more information about the impact factor, its limitations, and related statistics.
The abstract of this article provides an overview of this year's impact factors of the main journals in our disciplines, along with some new entrants. I was disappointed not to find the International Journal of Sports Physiology and Performance amongst them. This journal is now in its fourth year of publication and could have qualified for inclusion. I have therefore calculated its impact factor by putting its name and the years 2006-2007 into the advanced search form of Google Scholar. I retrieved 88 articles, and the total cites to these articles was 166. If we assume half the cites were made in journals published last year, the impact factor is approximately 1.0, which is a very good entry level. The value next year is likely to be lower, because 70 of the cites were to the article on magnitude-based inferences by Alan Batterham and me in the first issue, which will not be included in next year's calculation.
As in previous updates I will finish this article by introducing you to another citation statistic. Earlier this year a colleague brought my attention to the H index or number, named after the physicist Jorge Hirsch, who suggested it as a measure of an individual's lifetime publishing impact. Rather than attempting to define the H index, I will explain how I got it for a particular individual. I chose Tom Reilly, a founding father not only of British and European sport science but also of chronobiology and ergonomics, who died on June 11. I started by putting t-reilly into the author field of the Google Scholar advanced search form. The resulting reference list comes up pre-sorted in approximate descending order of the number of cites to each reference. I counted down this list from the top until I reached the last reference where the number of cites exceeded or equaled the number of the reference. The number of that reference is Tom Reilly's H index, 38, which means that he had 38 publications each cited at least 38 times. We can now regard this number as a benchmark for outstanding lifetime impact by a sport scientist. Reilly and Thomas (1976) was his publication with the most cites (186). His total number of publications is difficult to determine precisely because of the other T Reillys, but it appears to be ~500.
The procedure to get the H index was actually a little more difficult than I described above. To cut down on other T Reillys, I limited the search to references with at least one of the words exercise, sport, physical activity, ergonomic, and circadian. The list was still heavily contaminated with other T Reillys, most of which I eliminated by inserting tm-reilly wt-reilly dt-reilly tj-reilly te-reilly jt-reilly in the search field labeled without the words. The sorting was also far from perfect, so I copied the first 50 references into a Word doc to manually complete the editing and sorting, and I also skimmed the next 100 references for any highly cited publications that were out of sequence.
There are several conceptual problems with the H index: it discriminates against individuals who have been influential by publishing only a few very highly cited articles; its value will be lower when estimated with a conservative database such as Web of Science; and by being essentially non-parametric it does not adequately take an individual's total productivity (publications) or total influence (cites) into account. On the other hand, the H index is easy to calculate and verify, if your name isn't too ordinary or you aren't too productive. With these limitations, the H index is good for some fun of the mine's-bigger-than-yours variety, but there will be a need for serious validation and benchmarking if it is used to award grants and promotions. Find out more about the index in a Wikipedia article and in Hirsch (2005).
Hirsch JE (2005). An index to quantify an individual’s scientific research output. Proceedings of the National Academy of Sciences USA 102, 16569–16572
Reilly T, Thomas V (1976). A motion analysis of work-rate in different positional roles in professional football match-play. Journal of Human Movement Studies 2, 87-97
Thomson Scientific, Inc. is the
publisher and copyright owner of the Journal Citation Reports®.
Published June 2009