[This post is part of an ongoing series about impact factors]
The latest results from the annual impact factor boxing world championship contest are out. This is a combat sport where scientific journals are scored according to their supposed influence and impact in Science. This years competition rankings include the first-ever update to the newly introduced Five Year Impact Factor and Eigenfactor™ Metrics [1,2] in Journal Citation Reports (JCR) on the Web (see www.isiknowledge.com/JCR warning: clunky website requires subscription*), presumably in response to widespread criticism of impact factors. The Eigenfactor™ seems to correlate quite closely with the impact factor scores, both of which work at the level of the journal, although they use different methods for measuring a given journals impact. However, what many authors are often more interested in is the impact of an individual article, not the journal where it was published. So it would be interesting to see how the figures below tally with Google Scholar, see also comments by Abhishek Tiwari. I’ve included a table below of bioinformatics impact factors, updated for June 2009. Of course, when I say 2009 (today), I mean 2008 (these are the latest figures available based on data from 2007) – so this shiny new information published this week is already out of date [3] and flawed [4,5] but here is a selection of the data anyway: [update: see figures published in June 2010.]
| Journal Title | 2008 data from isiknowledge.com/JCR | Eigenfactor™ Metrics | ||||||
|---|---|---|---|---|---|---|---|---|
| Total Cites | Impact Factor | 5-Year Impact Factor | Immediacy Index | Articles | Cited Half-life | Eigenfactor™ Score | Article Influence™ Score | |
| BMC Bionformatics | 8141 | 3.781 | 4.246 | 0.664 | 607 | 2.8 | 0.06649 | 1.730 |
| OUP Bioinformatics | 30344 | 4.328 | 6.481 | 0.566 | 643 | 4.8 | 0.18204 | 2.593 |
| Briefings in Bioinformatics | 2908 | 4.627 | 1.273 | 44 | 4.5 | 0.02188 | ||
| PLoS Computational Biology | 2730 | 5.895 | 6.144 | 0.826 | 253 | 2.1 | 0.03063 | 3.370 |
| Genome Biology | 9875 | 6.153 | 7.812 | 0.961 | 229 | 4.4 | 0.07930 | 3.858 |
| Nucleic Acids Research | 86787 | 6.878 | 6.968 | 1.635 | 1070 | 6.5 | 0.37108 | 2.963 |
| PNAS | 416018 | 9.380 | 10.228 | 1.635 | 3508 | 7.4 | 1.69893 | 4.847 |
| Science | 409290 | 28.103 | 30.268 | 6.261 | 862 | 8.4 | 1.58344 | 16.283 |
| Nature | 443967 | 31.434 | 31.210 | 8.194 | 899 | 8.5 | 1.76407 | 17.278 |
The internet is radically changing the way we communicate and this includes scientific publishing, as media mogul Rupert Murdoch once pointed out big will not beat small any more – it will be the fast beating the slow. An interesting question for publishers and scientists is, how can the Web help the faster flyweight and featherweight boxers (smaller journals) compete and punch-above-their-weight with the reigning world champion heavyweights (Nature, Science and PNAS)? Will the heavyweight publishers always have the killer knockout punches? If you’ve got access to the internet, then you already have a ringside seat from which to watch all the action. This fight should be entertaining viewing and there is an awful lot of money riding on the outcome [6-11].
Seconds away, round two…
References
- Fersht, A. (2009). The most influential journals: Impact Factor and Eigenfactor Proceedings of the National Academy of Sciences, 106 (17), 6883-6884 DOI: 10.1073/pnas.0903307106
- Bergstrom, C., & West, J. (2008). Assessing citations with the Eigenfactor Metrics Neurology, 71 (23), 1850-1851 DOI: 10.1212/01.wnl.0000338904.37585.66
- Cockerill, M. (2004). Delayed impact: ISI’s citation tracking choices are keeping scientists in the dark. BMC Bioinformatics, 5 (1) DOI: 10.1186/1471-2105-5-93
- Allen, L., Jones, C., Dolby, K., Lynn, D., & Walport, M. (2009). Looking for Landmarks: The Role of Expert Review and Bibliometric Analysis in Evaluating Scientific Publication Outputs PLoS ONE, 4 (6) DOI: 10.1371/journal.pone.0005910
- Grant, R.P. (2009) On article-level metrics and other animals Nature Network
- Corbyn, Z. (2009) Do academic journals pose a threat to the advancement of Science? Times Higher Education
- Fenner, M. (2009) PLoS ONE: Interview with Peter Binfield Gobbledygook blog at Nature Network
- Hoyt, J. (2009) Who is killing science on the Web? Publishers or Scientists? Mendeley Blog
- Hull, D. (2009) Escape from the Impact Factor: The Great Escape? O’Really? blog
- Murray-Rust, P. (2009) THE article: Do academic journals pose a threat to the advancement of science? Peter Murray-Rust’s blog: A Scientist and the Web
- Wu, S. (2009) The evolution of Scientific Impact shirleywho.wordpress.com
* This important data should be freely available (e.g. no subscription), since crucial decisions about the allocation of public money depend on it, but that’s another story.
[More commentary on this post over at friendfeed. CC-licensed Fight Night Punch Test by djclear904]
O'Really? 
Does anyone actually read PNAS?
It is true, as I’m sure you know, that the number of citations reported by, say, ISI, and by Google Scholar, are never the same and can be radically different. This, of course, leads logically to the method some folks use: look up your article in both, and use the higher of the two numbers.
All these other factors though… I dunno what to make of them. Eigenfactors? Whazzat?
[visiting via Richard Grant’s blog at Nature Network, which I’m sure has an ENORMOUS Eigenthingy]
Comment by Richard Wintle — June 26, 2009 @ 4:49 pm |
Hi Richard, if I’ve understood it right, Eigenfactor is the Google PageRank algorithm applied to journals. Instead of all citations being counted as equal (regardless of where they come from) some citations are more important that others. E.g. if a paper is cited by an article in Nature (say) then that counts for more than a citation from the International Journal of Molecular Helminthology (say). This contrasts with the ISI Impact Factor where all citations are equal, at least, all the ones they decide to count anyway.
Comment by Duncan — June 26, 2009 @ 9:00 pm |
Another lucid critique of Impact factors is here – the point being that for specialist fields IFs overestimate the impact / influence of papers published in the general journals (as they “inherit” the big IF factor derived from the journal), and underestimate the influence of papers published in the specialist journals (where IFs are low).
Of course, even if one counts citations to/from individual papers, there are still major problems with inferring “quality” from this, especially in big fields. One reason is that a lot of readers/citers will still “access” the literature via review articles, so that a determining factor in whether a primary data paper gets cited comes to be whether it got cited in the better read reviews in the field. For an example of how this can distort the whole picture of the evidence see here.
Comment by draust — August 25, 2009 @ 4:22 pm |
Hi Dr. Aust, thanks for the references, they look interesting!
Comment by Duncan — August 28, 2009 @ 9:34 am |
[…] Impact Factor Boxing 2009 | O’Really? […]
Pingback by Science Spotlight: February 16th, 2010 | Next Generation Science — February 17, 2010 @ 5:16 am |
[…] I would be remiss not to mention Duncan’s “Impact Factor Boxing” article from last year, which was a large influence for this […]
Pingback by World Cup of Impact Factors | Vangelis Simeonidis — March 24, 2015 @ 11:59 pm |
[…] I would be remiss not to mention Duncan’s “Impact Factor Boxing” article from last year, which was a large influence for this […]
Pingback by World Cup of Impact Factors | V for Valence — March 28, 2015 @ 1:08 am |