The H-index was the brainchild of Jorge Hirsch. It's a method to quantify a researchers impact. (To read more on this, check this entry of mine: What's your h-index.)
I was just alerted that Michael Schwartzbach has written a program that can calculate your very own h-index quickly over the web by hacking into Google Scholar. I tried using it to get mine ... unfortunately "A Palazzo", is not unique and so I couldn't recalculate my own h-index. I finally got it to work by typing "A Palazzo" and Columbia and got 8 (the value I had the last time I checked). Go visit Epidemiologic Inquiry to use this neat web feature.
- Log in to post comments
The calculator isn't very good. It gave me a h of 9; I was pretty depressed, until I used the Web of Science directly, and found the actual number was 23.
How did you calculate it from Web of Science? the H-index is not as easy to determine manually as you think. It is from a natural equilibrium maximization function between total number of papers and the average number of citations over a large number of publications. One cannot just eye-ball it quickly.
One should read http://en.wikipedia.org/wiki/H-index to learn more.
I've run famous and low-ranking scientists and the feature seems to work okay. Try using your full name and middle initial- that works better than first initial. If one has worked/researched at multiple schools, it might be good to put both university names and departments in quotes, joined by an "OR".
An H-index value of 9 is actually rather high for a beginning scientist. you should be proud.
In epidemiology at least, while top scientists have H-index of >90, associate and assistant professors usually only have an H-index of 10 to 30. There are also full professors at harvard who have H-index of only 16 (though full profs generally average 40 or more in medicine).
Use web of science to pull up all your papers.
http://apps.isiknowledge.com/WoS/CIW.cgi?SID=E1aHKhlIeIkFaJllEN@&Func=C…
If you have a unique name+initials this is easy. If not, you have to sort a bit by institution. Then use the same web page to rank them by citation. Page down the rank until you come to the last paper whose number of citations is greater than its rank.
When you make a general search in web of knowledge you will get a list of publications. On the right side of the results page you will find a link to: "Citation Report". If you click it you will be led to a page where the h-ndex is displayed in the upper right corner. (I hope this works from any location and is not a special feature of our institution's ISI license.) The results are much more reliable compared to the link in your post, especially if you have a quite common name. If you were "Smith" additional information (like affilation) would't help much even in ISI. Then you'll have to mark all the papers that are actually yours which may be quite time consuming.
The competing database Scopus is trying to overcome this issue by adressing all authors with unique identifiers. Unfortunately, I was listed in the scopus database under several identifiers and they will only change it when the author sends them a list of all his publications.
BTW, the depth of calculating the h-index in ISI web of knowledge depends on the license your institution owns. Ours is limited to the years past 1996. Thus, if you published something before you should contact a colleague who has access to a full license covvering all years.
"I was just alerted that Michael Schwartzbach has written a program that can calculate your very own h-index quickly over the web by hacking into Google Scholar. "
Well, so long as it's not keeping him from important stuff like proof-reading my masters' thesis ...