How to use Web of Science to calculate your h-index
Web of Science is a database that has compiled millions of articles and citations. This data can then be used to calculate all sorts of bibliographic metrics including a h-index. Unfortunately, Web of Science is not free, but your institution might have a subscription to it. If so, you'll have access to it when you're on the campus network. When you're off campus, take a look at our off campus access database to see if your institution is listed.
The h-index was designed for evaluating authors, but it can also be used as a guiding metric when researching a new area. Technically, any set of papers can be given a h-index and it is fairly easy to apply it to a set of search results and use the h-index as a guideline for gaining a broad introduction to the major discussions in the field. But first let's explore the most efficient way to calculate the h-index of an author.
Step 1: The Web of Science search form
Go to: https://webofknowledge.com/. Don't worry, you're in the right place; Web of Science was recently rebranded and the domain hasn't changed as many people know it as Web of Knowledge.
Change the drop-down option from Topic to Author. The default search space, the Core Collection, is usually the best option as it includes both journals and conference proceedings. Note that the conference proceedings only go back to 1990. Changing the search space to a more specific domain database only makes sense if you get too many false positives in your search results and picking the right ones becomes cumbersome. If you or the author you are researching has a ResearcherID or ORCID entry then go with this ID and change the drop-down option to Author identifier.
If you are sure that the author's name is the same on all publications then provide as much information that you have because it will help in getting only relevant entries. Let's take for example the well known physicist Stephen W. Hawking. The problem is that he might go by Stephen Hawking, Stephen W Hawking, or just by SW Hawking. That's something that needs to be explored on an individual basis. In this tutorial, I try it with "Hawking SW". Yes, the last name goes first!
Step 2: Screening the search results and creating a citation report
The next step is to screen the search results if they really list the author we have been looking for. You can use various filters in the left hand panel to trim down your results. "Web of Science Categories" and "Organizations-Enhanced" are the most useful ones. But don't worry if some false positives are left we can remove them in the next step.
Once you have filtered the search results accordingly, all you need to do is click on "Create Citation Report".
Step 3: Assess the Web of Science h-index
By now, a new page will have been opened that displays all sorts of bibliographic metrics including the h-index.
Trouble shooting: what to do when there are articles that where wrongly assigned
At the bottom of the page we will find a list of all publications that have been included in this analysis. If you need to remove some false positive hits, then simply select the check-box of the entry and then hit the "Go" button. The h-index calculation will update immediately.
It's not just authors you can calculate a h-index for on Web of Science
Yes, the h-index was originally defined as an author-centric measure, but it can also be used to explore research topics in general. For example, if researching the topic of "ageism" on Web of Science, the h-index can be determined easily by following the steps described before and creating a citation report. For the search term “Machine Learning” the regular way does not work as the search yields more than 10,000 results. You can still calculate a h-index, but it requires some manual work. First change the "Sorted by" dropdown menu to "Times Cited -- highest to lowest", and then scroll to the point at which the times cited is less than the rank. Check out our h-index how-to for more infromation on how to calculate a h-index manually.