Tuesday, December 9, 2014

Impact factor

Another new impact factor by Scopus.

IPP: The IPP measures the ratio of citations in a year (Y) to scholarly papers published in the three previous years (Y-1, Y-2, Y-3) divided by the number of scholarly papers published in those same years (Y-1, Y-2, Y-3). 

SNIP: Source Normalized Impact per Paper measures contextual citation impact by weighting citations based on the total number of citations in a subject field.

SJR: SCImago Journal Rank is a prestige metric based on the idea that not all citations are the same.


Of course, you have the famous journal impact factor (JIF) by Thomson Reuters as published by Journal Citation Reports (JCR). This is based on 2 and 5 year windows. Thus, now we have impact factors on 2, 3 and 5 year windows but based on different databases.

The Eigenfactor is another PageRank-type measure of journal influence.

Scientists normally choose a journal based on factors such as longevity, publisher and reputation in the community rather than impact factor. However, in my experience, in the Indian scientific community, impact factor is given a lot of importance rather than how many times the paper is eventually read/cited. It should be remembered that less than 5% of the total papers that are published are cited more than 10 times (the i-10 index) and only 20% of the papers published in the journal contribute to 80% of the impact factor (Pareto principle).

4 comments:

Ankur Kulkarni said...

Another factor which I consider (may be other scientists also do) is the editorial board. One learns about the journal's emphasis and tastes based on its editors. This matters quite a bit when your work is somewhere between fundamental and applied.

Vijay Sethuraman said...

As an electrochemist, I prefer to publish in my society journals. I'm sure there are chemists who publish only/mostly in ACS/RSC journals (most of which are top journals), physicists who publish in APS journals, etc. You correctly point out the importance of longevity, and reputation in the community over impact factor. I wish more people recognized this. It appears to me that the impact factors and other such scientometrics are just marketing gimmicks. Mainly because publishing these days is big business and there is lot of money in it. Yet, bulk of the scientific and editorial work is done by hard-working graduate students, postdocs and faculty - usually from academic/research institutions. And the ironic thing is the libraries in these institutions pay big money to get access to this very same work! I wish every professor/scientist who publishes a paper shares their original content on the web for free.

Anonymous said...

Who cares about impact factor and citations?

Like they said in IISc senate meeting, i-10 and i-20 look like car models. Who cares if your paper has been cited ten times or cited at all.

The work should be relevant and useful. If Einsten work has not been cited, will it be lesser work?

Anonymous said...

Vijay,

Prof. Giridhar used to maintain the eprint site of IISc. That had the largest repository of all papers published from IISc, around 30000 papers were in the site.

Therefore, it was the best site in India where you could pick up all the published materials in IISc.

Do not what happened but it fizzled out when Giri left.


Anon@December 9, 2014 at 10:37 AM:

Citations are the only way to show that the paper has been read and the author found it useful. The number of citations for Einstein original work is nearly 90000. And spell Einstein correctly.

Senior IISc Prof.