The urban myth about h-index and ERC is that the h-index serves as a means for comparing applicants and assessing their scientific productivity and impact. Since an impressive publication list is the “bread and butter of a competitive ERC applicant”, many applicants tend to rely mainly on h-index as an indicator for their performance as scientists. Therefore, one may assume that if your h-index is not high enough you should not even consider applying to ERC. But is that really so?

What is the h-index? Let’s start with the basics.

Wikipedia defines h-index as “an author-level metric that attempts to measure both the productivity and citation impact of the publications of a scientist or scholar.”

It was introduced back in 2005 by Jorge E. Hirsch – a theoretical physicist, in order to measure the publication performance of researchers. Soon it became a popular index used by many. However, truth being – it is far from being an accurate way to compare and assess scholarly performance (see why below).

Since the ERC was established in 2007, scholars started quoting their h-index in their ERC applications, although never explicitly or officially required.

Why is quoting h-index problematic?

In the context of ERC, one should keep in mind that the h-index calculation method may give a misleading output. You can learn about the mathematics behind it here, and about the criticism here.

The main points being:

  • As it is a function of the number of publications and citations, the database used to extract this information strongly influences the index. The different databases (ISI, Scopus, Google scholar) may give very different results as their journal coverage and analysis methods are different. In fact, for an advanced scholar this can mean a difference of hundreds and even thousands of citations across databases. This is apparent when looking at results for one researcher, and becomes a critical issue when trying to compare researchers from different disciplines. Read more about this in a white paper by Anne-Wil Harzing here.  
  • The h-index suffers from a strong discipline bias. For example, it tends to be more indicative of scientific productivity and impact in life sciences, chemistry or physics.  In many of the engineering areas and most social sciences and humanities- it can be a very weak indicator. There are a few reasons for that:
    • Some disciplines tend to cite and publish more often than others.
    • The h-index does not take into account the relative contribution of a given author or the total number of co-authors of a given publication. Some disciplines tend to have more co-authors per paper. This may mean more papers per author albeit with lower relative contribution and less “research effort” in some of the papers.
    • For some disciplines, research output is typically presented as books or conference proceedings which are not well represented in the databases used for calculating the h-index.

The criticism about the inaccuracy of the h-index as a tool for comparing scholarly productivity and impact resulted in various proposals to modify it. Also, additional indices were introduced as alternatives. To name a few: i10-index (by Google), g-index, e-index, c-index, s-index, etc. Still, and despite its inaccuracy and bias, the h-index is the most popular performance index.

h-index and ERC

Clearly if you have a high h-index, you should use it in your ERC application. But what should one do if it is not as high?

The answer can be found in the ERC guidelines. This is the relevant instruction taken from the ERC work program: “…field-relevant bibliometric indicators may also be included”. This very short instruction gives you two clear answers, which are also in line with our experience in ERC:

  • You don’t have to mention the h-index or any other indices for that matter if they are not representative of your achievements. Mention these only if they add value to your presentation as a researcher.
  • Use only indicators that are relevant to your field. If h-index is not relevant to your field, don’t use it. Simply ignore it. If there is a more relevant alternative, use it. If not, don’t mention any indicator at all.

To sum it up: Don’t be afraid of low h-index. It may not be relevant in your case. Though some researchers are choosing to add this to their proposal, it has never been explicitly or officially required. For this reason, we suggest to use only when it is high. If h-index is not relevant to your field, do not mention it, and consider using a more relevant indicator. Have doubts or questions? We are here to help.

For additional consulting assistance for your ERC grant proposal, consider our Deep Dive service.

Supporting ERC applicants since 2007

Categories: ERC