At a recent meeting between the JSAP Associate Editors and the Journal's publisher, Wiley-Blackwell, the topic of conversation inevitably turned to impact factor.
When the 2010 impact factor for the Journal was announced there was a real sense of achievement that it had at last exceeded the magical figure of 1.0. Standing at 1.173 was not much higher than 1.0 but it most definitely exceeded it! Disappointment set in with the announcement of the 2011 impact factor. At exactly 1.0 there had been a small decrease over the year, but on the brighter side 1.0 was most definitely not less than 1.0 and perhaps it simply represented a degree of stability rather than a significant decline.
Despite various criticisms, the impact factor remains an important tool to compare journals within a given field. The impact factor of a journal is calculated as the total number of times its article are cited in the year by indexed journals, divided by the total number of citable items published by that journal during the preceding two years; citable items being scientific content such as original articles, reviews and case reports but not editorials, book reviews and non-scientific letters. First devised by Eugene Garfield in 1955, it has been the subject of several editorials for this Journal, most recently by Dunn (2006) and Mooney (2011).
Whilst the impact factor is not the driving force behind JSAP it would be imprudent to dismiss its importance. The higher the impact factor, the greater the number of submissions, potentially of more advanced scientific value, that the journal is likely to receive. But like any measurement the impact factor varies depending on counting of cites and calculation of citable articles, may be questioned with regard to consistency and reproducibility, and is open to editorial influence. That editorial influence may ultimately be beneficial for the given journal, may have limited effect on the type of submissions received and may not even be noticed by the readership. For instance, review articles, which evidence suggests are likely to be cited more than original research studies, can be encouraged and those papers most likely to be cited may be published early in the calendar year allowing more time for future citation. The number of articles unlikely to be cited can be limited and Early View (online publication ahead of print) encouraged as citations count but the articles do not form part of the denominator until published within a specific issue. Providing these influences have no negative effect on the Journal, JSAP takes account of them all. There has been an increasing number of review articles published recently with the ultimate aim of at least six annually. Many articles are available electronically weeks before print date with the ultimate aim of three complete future issues always available online. Case reports, although maintaining an important role, must comply with JSAP’s strict guidelines ensuring they contribute materially to the existing literature.
Certain methods used to influence impact factor can be questioned ethically. Self-citation can be abused and undue pressure may be put on authors to add spurious citations to an article (coercive citation) thereby inflating impact factor. An infamous manipulation was used by the specialist journal Folia Phoniatrica et Logopaedica in an effort to highlight the potential abuse of impact factor by publishing an editorial citing all of their articles from 2005 to 2006, thereby significantly increasing impact factor (Schuttea and Svec 2007). Of course, such nefarious practice may not go unpunished as future indexing in the Journal Citation Report® may be in jeopardy. With a reported self-citation index for 2011 of only 7% (184 of 2592 citations), JSAP is a long way from any such concerns.
Despite the small decrease in JSAP’s impact factor, there was consolation in a similar fate experienced by most other related and oft considered ‘competitor’ journals. JSAP moved from a ranking of 48 of 145 veterinary journals as listed by the Journal Citation Report® to 56, a position it held previously in 2010. By impact factor it is placed 14 of the 19 veterinary journals published by Wiley-Blackwell.
While impact factor provides a means of measurement and comparison for scientific journals, other methods continue to be devised for journals themselves and contributing authors including, amongst others, the H-index, immediacy index, Eigenfactor, cited half-life and the article influence score. But the biggest question, particularly for a generalist journal geared for practitioners such as JSAP, must centre on usage. Additionally, impact factor gives no clear indication on how useful or otherwise an article is to its reader and while JSAP insists that each article enunciate its own clinical significance, this is naturally biased by the authors themselves. With so much data now available online, we are beginning to get a flavour of the real possibility of measuring usage and perhaps even usefulness.
The Wiley Online Library was launched by the Journal's publisher Wiley-Blackwell in August 2010. In 2011, over 210 million full text articles from nearly 1,500 peer reviewed journals were downloaded from the Library, an increase of 39% compared with 2010, and a further increase is expected for 2012. Journal specific trends for JSAP are encouraging. Apart from August, traditionally a holiday period, over 16,000 unique visitors accessed the Journal each month in 2011, increasing to a monthly average of over 20,000 for 2012. A significant 30% of traffic now emanates from the USA compared to 15% from the UK, highlighting the global infiltration of JSAP today. In terms of number of downloads, JSAP is one of the top three veterinary journals published by Wiley-Blackwell. From January to October 2012 there were nearly 250,000 full text downloads representing an overall approximate increase of 20% on 2011. A comparison of the value used to calculate impact factor with full text downloads is astounding. In February 2011, Whitley and Day published a review article on immunomodulatory drugs. It has received eight citations to date, ranking in the top three for cited articles published in JSAP in 2010–11. Yet a staggering 6,500 full text articles were downloaded for the same time period. Those cynics amongst us may question whether such a figure truly represents usage or usefulness. It is admitted that review articles are free to download, eliminating a barrier that might otherwise be present. Many of us are skeptical of anything free of charge, having a sense that it is inherently devalued. Receiving a book free of charge may not inspire us to read it as avidly as when paid for. It is easier to pass it off as a gift to someone else, to use it to prop up a piece of furniture or to simply fill a gap on one's bookshelf. However, I don't recall ever hearing from the computer savvy I know (i.e. my children) about accessing free full text downloads simply because they are free only to become part of an unused collection. Surely from such a number of downloads, there is a significant proportion of use for the Whitley and Day (2011) article. There is no doubt that more usage-based data will become available in the future and article- and journal-level metrics will develop from this. Usefulness remains a more elusive factor to measure but will undoubtedly be a more reader dependent factor – it will be interesting to see how publishers and others rise to this challenge.