Talk:Journal ranking

From Wikipedia, the free encyclopedia

	This article is within the scope of WikiProject Academic Journals. To participate, you can edit the article. You can discuss the Project at its talk page.
???	This article has not been assessed

Copied from the Library article talk page:

[edit] eigenfactor

eigenfator.org (EF) [1] [2] [3], similar to the h-index, and a free alternative to the impact factor of the Journal Citation Reports, hasn't been scientifically evaluated yet, has it? -- Cherubino 17:44, 19 October 2007 (UTC)

this is work by Carl Bergstrom, a well-known academic. the website mentioned includes some citations to the literature. Most of the academic discussion at this point is on the SIGMETRICS listserv, and links to additional references as they are published will be found there. Someone should write the article--I will if nobody else does. Basically, eigenfactor modifies impact factor by first taking account of the importance of the journals being cited from in the manner of Google PageRank (this gives what Bergstrom calls "article influence" ), and then normalizing across disciplines to reflect the relative density of citations. (As I hope everyone using them realizes, impact factor comparisons are valid only within individual specific scientific fields.)

H index is actually something rather different in principle--it's a ranking by number of articles containing a particular number of citations, (h=20 means someone has published 20 articles each having 20 or more citations to it) and can be use for individual authors. Impact factors and eigenfactors apply only to collections of articles such as journals--h-factor for comparing journals is altogether controversial. It certainly is a very convenient single number for rating scientists, but it applies only within a particular very specific field. The actual validity for anything remains quite disputed. (I have somewhat of a bias, though--I think it measures productivity more than creativity). I've used it as a rough-and-ready argument at AfD, but I would not cite it in an article. (note that h indexes based on Google Scholar are really unreliable, having all the problematic factors of Google Scholar, such as the unknown criteria for inclusion and the erratic coverage)

The real question is probably whether eigenfactor is sufficiently accepted to be cited here. I think this should be done very gingerly at present, and certainly only in argument, not articles. For one thing, it will be necessary to explain them every time they are used.

anyone wanting impact factor comparisons in a particular field, just ask me by email. DGG (talk) 21:28, 21 October 2007 (UTC)

[edit] COI

This article was entirely written by the developer of the site journalranking.com., much discussed in this article. although it could reasonably be speedy deleted as advertisement, I think it would be more useful to convert it into a more general article. I welcome attempts at this, or I'll give it a try. DGG (talk) 02:48, 23 October 2007 (UTC)