|
|
Journal CiteScores
2017-2019: citation metrics from the Scopus
database Will G Hopkins, Institute for Health
and Sport, Victoria University, Melbourne, Australia. Email. Sportscience 25, i, 2021
(sportsci.org/2021/inbrief.htm#citescores. Reviewer: Andrew Stewart, Victoria
University, Melbourne, Australia. Published May 2021. ©2021 Updated 6 July. The spreadsheet
sorted by cite score had not been properly sorted. It now shows International Review of Sport and Exercise
as the journal with the highest score, followed closely by British Journal of Sports Medicine and
Sports Medicine. Updated 28 June. A new workbook shows the latest (2020) and only the previous year's (2019) cite scores, as published by Elsevier in June 2021. Amongst the top journals, Sports Medicine has made the biggest jump and is threatening British Journal of Sports Medicine for first place. My previous summary workbook showing the scores for 2017, 2018 and 2019 (and the comparison with the impact factors for the top-10 journals in 2019) is still available here and from the links in the item below. Updated 18 May to include relevant journals with measurement or musc in the title and a few more relevant clinical journals. Note: Frontiers in Sports and Active Living started publishing in 2019, so it is not yet in the Scopus or Web-of-Science databases. Download a workbook of the current year (2019) and two previous years of CiteScores for journals in sport and exercise medicine and science. Please email me with any journal titles I have missed and I will update the workbook. The rest of this item explains and compares the CiteScore and the traditional impact factor. After a four-year hiatus, I am again providing a summary of citation scores for the journals in the disciplines of exercise and sport science. As noted in my 2015 article, I abandoned the traditional impact factor in favor of Elsevier's metric, which is derived from a bibliographic database (Scopus) more relevant to sport and exercise science. The Scopus site allows free access to scores for individual journals and for journals grouped by subject area, but exercise and/or sport science are not included in the available subject areas. I have therefore downloaded the very large and unfriendly workbook (25 MB) of over 40,000 titles from the Scopus site and extracted our journals into a more user-friendly smaller (74 KB) workbook, which has spreadsheets sorted by 2019 CiteScore and by journal title. I have included on one of the tabs the SAS program I used to filter for our journals. The method of calculation of the CiteScore has changed somewhat since my last article; specifically, and to quote from the Scopus site, "[the] CiteScore [for a given serial for] 2019 counts the citations received in 2016-2019 to articles, reviews, conference papers, book chapters and data papers published in [the given serial in] 2016-2019, and divides this by the number of publications published in [the serial in] 2016-2019." The traditional impact factor (compiled by Clarivate, formerly Thomson-Reuters and Web of Science) is calculated from the citations in all articles published in 2020 to articles published in a given journal in 2018 and 2019. For formulae that make the definitions clear, see the Wikipedia articles on the CiteScore and impact factor. Elsevier's previous metric, the impact per paper, was calculated in the same manner as the impact factor, except that it used the previous three years rather than the previous two; the values were very similar, as detailed in my 2015 article. I am unaware of any comparison of the new CiteScore with the impact factor for our journals, so I accessed (via my institution) the impact factors of the top 10 journals in the sport-sciences category at the Clarivate site and added them to the spreadsheet. You will see that Clarivate has omitted many high-scoring journals relevant to our disciplines. You will also see that the CiteScores are higher than the impact factors, on average by 52%, with a range of 21% to 94% (a geometric mean factor of 1.52, with a range 1.21 to 1.94, as shown on the last tab of the spreadsheet). The discrepancies are much greater than those with the original impact-per-paper metric. Is the CiteScore better than the impact factor? Who knows? I suspect that neither is a particularly good measure of a journal's quality, but both are obviously good measures of a journal's popularity with researchers and prestige with bean counters. ———– |