Research metrics

The h-index is a relatively simple and straightforward indicator, which is calculated so that a researcher who has at least h papers that each have at least h citations has an h-index of value h. It intentionally ignores the total number of papers or total number of citations, so a very long tail of uncited papers will have no effect, and neither will a single exceptionally highly cited paper.

The value of a "normal" h-index varies significantly between fields, since it is based on the number of papers published, but in most fields 50+ would indicate a well-established senior researcher - they have at least fifty papers with at least fifty citations apiece. H-indices above about 70 are generally uncommon.

The h-index has become popular due to the simplicity of its calculation and its percieved "neutrality", but in practice it has a number of significant issues.

• It does not take account of the differences between fields - an h-index of 40 would indicate something very different in mathematics versus genetics, given the different rates of publication, levels of collaboration, expected levels of citation, and so on.
• It is very dependent on past history - the h-index is heavily driven by earlier papers. It thus can unexpectedly penalise researchers who took a career break a decade ago, or spent time teaching or in industry, even if their current publication record is excellent.
• Similarly, it is relatively unaffected by recent inactivity - for many established researchers, their h-index would not show a significant dip if they had not published anything in the past couple of years

We strongly recommend that you avoid using the h-index to directly assess individual researchers.

For any kind of internal assessment at UCL, this is one of the explicit requirements of the bibliometrics policy:

(6.) Avoid applying metrics to individual researchers, particularly metrics which do not account for individual variation or circumstances. For example, the h-index should not be used to directly compare individuals, because the number of papers and citations differs dramatically among fields and at different points in a career.

The h-index should also not be used for analysing larger groups, such as a department or an institution - when applied to larger groups of people, it tends simply to reflect the total number of papers published, and does not tell us anything informative.

Various alternative types of the h-index are regularly proposed, to address particular issues with the original indicator, but these often suffer from the same fundamental problems. None of them are widely accepted and we would not recommend them.

Ultimately, directly comparing the publication record of researchers is complicated and nuanced. It cannot and should not be reduced to a single numeric indicator.