Ways of Ranking Journals

January 18 to January 22, 2009 Summary: A variety of methods to rank journals are mentioned briefly including Weighted Impact Factor, PageRank, Eigenfactor, h-index, Index Copernicus, SCImago journal rank, F1000 Factor, and several other tools. Several links are provided. –MW.

Is there any scale or score to evaluate the medical journals other than impact factor?

Mohamed Reda Bassiouny
________________________
Yes, there are other ways to rank journals. Related to the Impact Factor is the Weighted Impact Factor, which weights the prestige of the cited journal (that is, whether its impact factor is higher or lower than that of the citing journal) when calculating the impact factor. The Weighted Impact Factor gives slightly different rankings than does the Impact Factor.

Two other citation metrics are the PageRank for electronic journals and the Eigenfactor for print journals. A PageRank is created by counting the number of "incoming links” to a specific Web site, such as that of a journal. Each link counts as one “vote.” In addition, each link is weighted by the importance of the source of the link: A page that has links from several pages with high PageRanks receives a high rank itself. If there are no links to a Web page, there is no support for that page. Thus, the higher the PageRank, the more influential the journal is believed to be.

The calculation of the Eigenfactor is similar to that of the PageRank, but instead of counting links to Web pages, it follows citations from one source to another and thus considers an entire network of sources. Citations are not limited to those from other scientific journals, as is the case with Journal Citation Reports, but include citations to newspapers, magazines, theses, technical reports, and so on, to better assess the impact of a given publication. Eigenfactors are scaled so that the sum of the Eigenfactors of all journals listed in Journal Citation Reports is 100.

The Eigenfactor is also used to calculate an article influence score, which is a measure of the average influence of each of its articles over the first 5 years since publication. Article Influence measures the average influence, per article, of the papers in a journal and is similar to the Impact Factor. Article influence scores are normed so that the mean article in the entire Journal Citation Reports database has an article influence of 1.00. The article influence score for the British Medical Journal in 2006 is 3.2874, meaning that the average BMJ article has a little more than 3 times the influence of the average article indexed in Journal Citation Reports.

Tom Lang
________________________
The following copy-paste from our recent editorial http://cytojournal.com/article.asp?issn=1742-6413;year=2008;volume=5;issue=1;spage=15;epage=15;aulast=Shidham may be helpful- The article also show the figures of snap shots some sites for quick reference.

3B. Real-time quality indicators available through FREE online open resources

As mentioned above, currently there are many other real-time, more flexible quality indicators allowing comparison between different journals in open system. These indicators not only provide quality evaluation of a particular journal, but also permit quantification of isolated articles and individual authors for comparative evaluation. The department chairs-leaders, academic institutions, readers, and authors have FREE access to monitor and quantify individual articles, journals, and authors for promotion and other academic purposes.

i. h-index: Sometimes referred to as the Hirsch index or Hirsch number, quantifies scientific productivity and impact by the author. An author with an index of h has published h articles. Each of these h articles has been cited by other scholars at least h times (http://en.wikipedia.org/wiki/Hirsch_number#cite_note-2). This index is useful for comparing authors working in the same field as citation patterns differ widely among different fields. This index is ideal for promotion and comparison of different faculty in the same area of expertise.

ii. Index Copernicus (http://journals.indexcopernicus.com/info.php) provided through the freely available web site, which also explains how the index is generated (http://journals.indexcopernicus.com/info.php) with opportunity to compare with other journals from the data about other journals under the 'Master List' http://journals.indexcopernicus.com/masterlist.php?litera=aandstart=0andskok=30

iii. SCImago research group (http://www.scimagojr.com/index.php) has numerous features including SJR, H-index, and other quality indicators for comparison of various journals as a free resource.

iv. Harzing.com provides free software for citation analysis of individual authors and journals. To download this software visit http://www.harzing.com/resources.htm and click on 'Publish or Perish installer for Windows' (or for Linux) towards the bottom of this page by scrolling downward [Figure 3] Citation analyses can be performed for individual authors [Figure 4] or journals [Figure 5] in 2 minutes based on online real-time Google scholar data http://scholar.google.co.uk/

Vinod B Shidham
________________________
Tom Lang and CytoJournal editor explanations have been quite useful. In addition, I suggest you get the following papers, which deal with this particular and hot topic. Besides papers on traditional impact factor and H index, you will find recent information on the European Factor, the single researcher impact factor, and, more recently, the personal impact factor by Graczynski.

I hope it could be helpful.

Daniel Limonta
Associate Editor, Journal of Infection in Developing Countries (JIDC)

Seglen P. Why the impact factor of journals should not be used for evaluating research. BMJ 1997;314(7097):498-502.

Garfield E. The history and meaning of the journal impact factor. JAMA 2006;295(1):90-3.

Editors. The impact factor game. Plos Medicine 2006;3(6):707.

Dong P, Loh M, Mondry A. The "impact factor" revisited. Biomed Digit Libr 2005;2:7.

Hofbauer R, Frass M, Gmeiner B, Kaye AD. The European Factor – The Euro-FactorTM. The new European Journal Quality Factor. The new European “scientific currency”. Vienna: VICER Publishing; 2002.

Graczynski MR. Personal impact factor: the need for speed. Med Sci Monit 2008;14(10):ED1-2.

Burnham JF. Scopus database: a review. Biomed Digit Libr 2006;3:1.

Hirsch JE. An index to quantify an individual's scientific research output. Proc Natl Acad Sci U S A 2005;102(46):16569-72.

Hirsch JE. Does the H index have predictive power? Proc Natl Acad Sci U S A 2007;104(49):19193-8.

Castelnuovo G. Ditching impact factors: Time for the single researcher impact factor. BMJ 2008;336(7648):789.

Castelnuovo G. More on impact factors. Epidemiology 2008;19(5):762-3.

________________________
Recently, we also proposed a new metric for evaluation of journals, and named it "Weighted Impact Factor." For calculation of this metric, unlike the traditional Impact Factor, which gives a weight of 1 to all citations, we give a weight between 0.1 and 10 to each citation, depending on the prestige of the citing journal relative to the cited journal. The source is Journal of Informetrics 2008 April; 2(2): 164-172. The DOI is 10.1016/j.joi.2008.02.001

The abstract of this article is available from http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B83WV-4S7HSJF-1&_user=10&_coverDate=04%2F30%2F2008&_rdoc=7&_fmt=high&_orig=browse&_srch=doc-info(%23toc%2333796%232008%23999979997%23685880%23FLA%23display%23Volume)&_cdi=33796&_sort=d&_docanchor=&_ct=7&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=8dd2390889eecec63de6b2ae80415d4d

For those who are interested in reading the full text of the article, please contact me directly for the PDF version of the reprint.

Farrokh Habibzadeh
________________________
There is also the F1000 Factor where members of Faculty of 1000 Medicine and Biology rate the significance of selected articles to help readers focus their reading. By rating articles based on their individual merits rather than the journal in which they appear, this system provides a useful alternative to impact factors and citation scores.

Faculty of 1000 Medicine is available at www.f1000medicine.com

Faculty of 1000 Biology is available at www.f1000biology.com

I should mention that I am MD for Faculty of 1000 and have been involved with these two services since their launch in 2002.

Other scales are Sciimago available from Scopus (http://www.scimagojr.com/), Google Scholar citations (http://scholar.google.co.uk/) and the h index (available via http://www.harzing.com/pop.htm#whatfor), among others.

Anne N Greenwood
________________________
Thank you for your valuable answers regarding journal ranking. I appreciate your inputs.

Mohamed Reda Bassiouny
________________________
This concept of "Weighted Impact Factor" is interesting. Unfortunately, I did not have access to the full text of your article to find out what numerical value you used as an indicator of the concept of 'prestigious' as you stated in the abstract. I think one way is to use IF ratios (Citing Journal If / Cited Journal If) to obtain the prestigiousness. This ratio may then be used directly as a floating point coefficient for the apparent journal IF or truncated to a more meaningfull integer before applying it to the IF.

Based on this Weighted Impact Factor it may be difficult to predict what will happen after several rounds of applying it. What will happen to the IF of most prestigious journals? Based on this formula, such journals will have little room for increasing IF, because the more prestigious the journal, the less will be the number of potential citing journal with higher prestige than it. Therefore, these prestigious journals will be cited mainly by less prestigious ones, which in turn lower their IF. In the long run, it may have an equalizing effect, but, to be sure, some simulation work may be needed.

Mahmoud Saghaei
________________________
I will send you the reprint of the article. Indeed, we used the quotient of previous year impact factors of the citing and cited journals as an estimate for the relative prestigious of the citing journal relative to the cited journal. However, since this quotient would vary from zero to infinity (based on the IFs of the citing and cited journals which for general medical journals varies from almost 0 to almost 50), we developed a logistic function by which we normalize that quotient. That normalization function has two main properties: It maps all the values between zero and infinity (the quotient) to an interval of 0.1 and 10, and it maps the quotient of 1 to 1 (when the IFs of both citing and cited journals are equal). For more details please refer to the article.

You're right that the variability of WIFs is less than the variability in IFs. We explained it in the article; indeed, IF has a very wide and skewed distribution, while WIF has a distribution much closer to normal distribution. You're also correct that using WIF, the high ranked journals could not keep their high rank, if they want to continue their work as they have done before. As you know, it's not much difficult to be the first one; the most difficult thing is stay the first one. For these high ranked journals, for being high ranked, it is not only necessary to receive citations from other journals, but they should also publish better articles to receive more citations. WIF indeed reflects both the popularity (number of citations) and prestige of a journal (which is taken into account in the calculation of weights). Furthermore, we propose that for the next rounds of calculation, it is better to consider the quotient of WIFs of the citing and cited journals as estimates of relative prestige.

F Habibzadeh
________________________
PLoS ONE announced an "articles-level metrics project" in which it will release various types of data on citations, page views, press coverage, etc for individual articles. The idea is to let readers decide what is important and to dilute the importance of any one metric of a paper or a journal.

Sections

Personal tools

Ways of Ranking Journals

Document Actions