Skip to main content

Bibliometrics

Indicateurs bibliométriques
Bibliometric indicators provide a quantitative appraisal of periodicals, published texts or researchers.

Bibliometrics are primarily calculated on the basis of citations from a corpus of scientific articles and are mainly used in certain disciplines (science, medicine, etc.). They can only ever be complementary to qualitative assessments. They have been the subject of various criticisms, in particular when their use goes beyond their intended purpose.

They may be useful for:

  • Identifying the impact of scientific publications (primarily scientific articles).
  • Identifying periodicals that are considered to be important in your field of research (so you can read them or publish in them).
  • Assessing the reputation of researchers, laboratories, universities, countries, etc. on the national or international scientific scene.

There are alternative measures (‘Altmetrics’), which use, among other things:

  • Shares on social media (Twitter, Facebook, LinkedIn, etc.).
  • Downloads on platforms like Mendeley.
  • Mentions in blogs or wikis.

Indicators

This is a selection of the most frequently used indicators.

1 - Article's impact

The number of citations measures the number of times a document has been cited by other scientific publications, within a given corpus and over a given period of time. This number, therefore, differs depending on the databases (Web of Science, Scopus, Google Scholar, PubMed, etc.), which do not have the same scope.

Measuring the use of articles on the Internet, and, in particular, via social networks, may be a useful way to supplement the number of citations in a given bibliographical database. More information

2 - Periodical's impact

Thomson Reuters’s Impact Factor© provides an indication of the impact of a periodical and represents the ratio between the number of citations a periodical’s articles receive over a given period of time (two or five years), based on the Web of Science corpus, and the number of articles published in this same periodical during the same period. The IF has been the subject of various criticisms, but remains broadly used in certain disciplines. There are two versions of the IF: IF Science and IF Social Science. Impact Factors are published each year in the Journal Citation Reports

More information

The EigenFactor assesses the influence of a periodical, by quantifying the number of times when articles from a periodical that was published within the past five years have been cited in the reference year, based on the Web of Science corpus. Furthermore, the EigenFactor weighs each reference according to a stochastic measurement of the time spent by researchers consulting a particular periodical. The Article Influence Score measures the average influence of each of the articles from a periodical over the five years following its publication.

Eigenfactor.org

The SJR indicator (now the SJR2) assesses the ‘prestige’ of a periodical based on the number of citations, by applying an algorithm similar to Google’s (PageRank) to the articles from a periodical, which are indexed by the Scopus database. This is based on the idea that ‘not all citations are equal’, in particular depending on the field of research.

Scimago Journal & Country Rank

Journal Metrics

Scopus (intranet- or VPN-access only)

The SNIP indicator (now the SNIP2) assesses the influence of a review based on the number of citations by weighing them based on their context (field of research), based on the Scopus corpus.

CWTS Journal Indicators

Journal Metrics

Scopus (intranet- or VPN-access only)

The IPP indicator provides an indication of the impact of a periodical and measures, on the basis of the Scopus corpus, the ratio between the number of citations that the articles from a periodical has received over the past three years and the number of articles published in this periodical during the same period. Unlike the SNIP and the SJR, the IPP is not standardised per field of research.

CWTS Journal Indicators

Journal Metrics

Scopus (intranet- or VPN-access only)

The h-index (see below) may also be used to assess a review.

3 - Researcher's (or group of researchers) impact

Named after its creator, Jorge H. Hirsch, the h-index assesses the scientific output of researchers based on citations within a given corpus. An h-index of N means that the researcher in question (or the group of researchers) has published N articles, which have been cited at least N times in the database in question. Therefore, different h-indexes exist for the Web of Science, Scopus, Google, ORBi, etc. The h-index may be used to measure the scientific activity of a university, country, etc., as well as the activity of a periodical.

The G-index is a variation of the h-index. It corresponds to the ‘largest number of articles for which we can say that all the g articles have received at least g² citations’.

This index favours researchers who have received a very large number of citations for some of their articles.
For example, an author has a G-index of 20 if they have published at least 20 articles, which, together, have received at least 400 citations.

The G-index depends on the corpus of references being considered and will be different depending on the database in question. Publish or Perish software makes it possible to automatically calculate a G-index based on the references of an author found on Google Scholar.

G-index (in french)

Publish or Perish

Last updated on