Search keywords
Very recently I installed a new plugin on this weblog.  A plugin is a standard way of extending the functionality of  WordPress software for self-hosted blogs. Our blog uses the WordPress package and is not hosted by the big providers as Blogger and WordPress, but is hosted by ourselves.

The new widget in the sidebar on the right, that has as title Search Keyword Cloud, is the result of  this new plugin. The new piece of software analyzes all visits that come to our site through search engines as Google or Bing.  This keyword analysis is possible because the search term is present in the headers of the http request to our server. The most frequent keywords are shown in the widget, not as a dull linear list, but as a cloud. The idea of a cloud, a web 2.0 concept,  is that viewers see quickly what the most important items are. The keyword search data are regularly reset to allow for changing search trends to show up more quickly.

The same keyword information is also available on pages of the admin panel of our blog. But that admin page is not visible to our authors, leave alone to outside readers. With the plugin installed we show the data to everybody, and in an attractive format.

I think this widget is an instantaneous success. Just look at it. You realize immediately what our readers are interested in: the majority of our viewers is interested in impact factors and more specifically in a list of impact factors. I never realized that journal impact factors are so important for early career scientists.

Impact factor
In the rest of this post I pretend to take citations seriously. In reality I do not take them seriously myself, but many others do.

What is an impact factor? The impact factor of a journal (IF)  is a measure for how often on average a paper in that journal gets cited in other scientific journals. It is calculated as follows:

The IF varies over time and in 2008 the IF of the journal Science was 28.1 and of the journal Nature 31.4 and the IF-leading journal is CA: A Cancer Journal for Clinicians that has an IF of 74.6. Owners and editors of journals with high impact factors like the impact factor a lot. High IF factor means a lot of attention, possibly high subscription rates and many submissions.

The IF factor is an invention of Thomson Reuters a for-profit company, that is in this business to make its share holders happy. The detrimental impact of their impact factor on science is of no concern to the company. The term impact factor and its abbreviation are not protected as brand names and can be used by other companies as well. Subscribers to Elsevier’s Scopus database can easily calculate similar factors but the Thomson Reuters impact factor is by far the most popular impact factor.

The way Thomson Reuters calculates the IF  (what is a citable item?) is a trade secret and publishing lists of the impact factors originating from their database is forbidden due to copyright protection and this right is aggressively defended by Thomson Reuters. You bet they read this post carefully.

How to get a list of impact factors
As I already outlined Thomson Reuters impact factors are protected by copyright. I have access to the Thomson Reuters database I can within a minute calculate all impact factors our viewers are interested in, but I cannot publish them here without getting into legal problems. The rich academic institutions all have a subscription to Thomson Reuters’ ISI Web of Science, but I have no idea how this is in the rest of the world.

Some people have the courage to publish lists of impact factors anyway. A more reliable way is through the web sites of publishers. Every journal publishes its own impact factors on their web site. So if you are in biology or chemistry, within half an hour you can have collected the relevant impact factors for your field.

If you type in “[organization] journal impact factors” to Google or Bing,with [organization] a learned society or publishing house you will immediate get their list. I give you five examples:

American Chemical Society
American Institute of Physics
American Physiological Society
Institute of Physics
Optical Society of America

Why do people want to know the impact factor?
The argumentation is as follows. To get a job as a researcher at an established institute you must have a good citation score (that is high h-index). A paper published in high IF-journals gets better cited, so if I want more citations I should publish my paper there. When I have a choice between a number of journals that are suitable to publish my manuscript I will want to publish my paper in the one with the highest IF.

These arguments have several flaws, of which I will mention a few. Your number of coauthors is important. Your position in the list of authors is important. These aspects are not in the IF. In addition the IF is an average number and the individual paper could get much different citation numbers.

I will give an example: the journal Nature has a large IF. So people want to publish in Nature, but for physicists this does not make much sense as the physics papers in Nature do not profit from this large impact factor.

There are a number of reasons for not submitting to a high IF journal. These journals have a very large rejection rate. It takes a large effort to get your paper in an acceptable format for their submission process. If your paper gets subsequently rejected, what is very likely, all your efforts were in vain. The paper has to be rewritten and you have a substantial delay on your first-discovery claim.

Impact factors can be manipulated
There is quite some criticism on the concept of the IF. Here is one article from a Plos editor. In a recent, very informative, article the American Physical Society (APS) indicates how impact factors depend very much on the number of papers a journal publishes. The high-impact journals publish orders of magnitude less papers than the more specialized professional journals. The APS shows that any publisher publishing more than 500 physics papers per year cannot get an IF higher than 20.

In addition the IF can be manipulated. There are persistent rumors about companies negotiating with Thomson Reuters to change the numerator or the denominator in the formula for their journals IF. As the whole IF calculation by Thomson Reuters is hush-hush the rumors could well be true.

Journals adjust their policy to increase their IF. Review-type papers help to increase the IF. And indeed the number of reviews published by journals have gone up tremendously. Kai Simmons has recently strongly criticized in the journal Science (also available directly from his own website) the use of the IF and the ways it can be manipulated.

The recipe for a new journal to get a very high IF is as follows. Try to be as narrow as possible. Publish as many reviews as possible. Publish as few papers as possible (CA: A Cancer Journal for Clinicians publishes less than 50 articles per year). Get well-established scientists to write solicited comments in your journal describing how important a particular article in the present issue is.

My PhD students decide
How do I make the choice where to publish? To be honest I give my PhD students my advice, but finally they decide. Their career is at stake. So if they want to submit to Nature, we submit to Nature. And after the paper is rejected we rewrite it and then publish somewhere else.

Getting more citations
Does it help  for early-career scientists to have their paper published in a high-impact journal? To get a job it certainly helps. For getting more citations the content is much more important.

If you are concerned with increasing your citation score it makes more sense to submit your paper not to the top-IF journals but to a large mono-disciplinary journal with a not unreasonably low IF. In physics the Physical Review is a good example. The Journal of the Optical Society of America is another one. These journals have reasonable rejection rates and are well read by the community.

So eager to submit
I notice that after enough results have been obtained young scientists often want to submit their manuscript with these new results quickly, and of course in a high-impact journal. These authors invariably underestimate how much they can improve their manuscript if they spend more time on it. I never understand why people are so reluctant to spend more time on their manuscript concept. About 1/3 of my book is dedicated how to improve your scientific papers and as a consequence increase your readership.

Follow