h Index

The h index is a bibliometric indicator that measures both the productivity and citation impact of a researcher, group, or journal. A researcher has an h index of h when h of their published papers have each been cited at least h times and the remaining papers have no more than h citations each. For example, an h index of 25 means the researcher has 25 papers that have each received at least 25 citations, but not 26 papers with at least 26 citations each. The metric was proposed by physicist Jorge Hirsch in 2005 and has become one of the most widely recognized indicators of scholarly impact.

Before the h index, research impact was typically assessed using either total publication count (which rewards productivity regardless of quality) or total citation count (which can be inflated by a single highly cited paper). The h index combines both dimensions into a single number, capturing the idea that an impactful researcher produces a body of work that attracts sustained attention. It has been adopted by major academic databases (Google Scholar, Scopus, Web of Science) and is used, formally or informally, in hiring, promotion, tenure, and funding decisions. While it has significant limitations (discussed below), its simplicity and intuitive interpretation have made it one of the most widely used research metrics in academia.

To calculate the h index, a researcher's publications are ranked by their citation count in descending order. The h index is the largest number h such that h papers have at least h citations each. Graphically, it is the point where the citation curve crosses the line y = x (the rank).

Computed on a growing citation record, the h index can only increase or stay the same over a career, because citations accumulate. A reported value can still fall when a database corrects its records or self citations are excluded, and windowed variants such as the h5 index can fall as older papers leave the window. The h index increases slowly for researchers who publish many low impact papers and for researchers who publish few papers, even if those papers are highly cited. The h index tends to be higher in fields with larger research communities and higher average citation rates (such as biomedical sciences) than in fields with smaller communities (such as mathematics or humanities).

Several variants and extensions have been proposed. The h5 index considers only publications from the last 5 years, capturing recent impact. The g index gives more weight to highly cited papers (a researcher's g index is the largest number g such that the top g papers have collectively received at least g squared citations). The i10 index (used by Google Scholar) counts papers with at least 10 citations. The m quotient divides the h index by the number of years since the first publication, adjusting for career length.

The h index is database dependent: a researcher's h index may differ between Google Scholar, Scopus, and Web of Science because these databases index different sets of publications and count citations differently.

The h index is most useful as one of several indicators in a portfolio approach to research evaluation. It can be used to compare researchers within the same field and at similar career stages, to track the impact trajectory of a researcher over time, or to provide a quick summary of a body of work. It is reported by the major citation databases (Google Scholar, Scopus and Web of Science) and is easy to compute. It should always be supplemented by other metrics (such as field normalized citation counts, qualitative assessment of key publications, and peer review) and interpreted in context.

A common mistake is using the h index as the sole criterion for evaluating research quality or making hiring and funding decisions. The h index has well documented biases: it favors researchers in large, high citation fields over those in small or low citation fields; it disadvantages early career researchers (who have had less time to accumulate citations); it does not account for the number of authors on a paper (a middle author on a highly cited collaborative paper benefits equally); and it does not distinguish between self citations and independent citations. Comparing h indices across different academic disciplines is misleading without normalization. Another mistake is treating the h index as a measure of the quality or importance of individual papers, when it is a measure of a body of work. Finally, manipulating the h index through excessive self citation, citation rings, or salami slicing (dividing work into many small publications) corrupts the metric and is widely regarded as a breach of publication ethics.

A mid career epidemiologist has published 85 papers over 15 years. Ranked by citation count, the 42nd most cited paper has 48 citations and the 43rd has 41 citations. The researcher's h index is 42 (42 papers with at least 42 citations each). The h5 index (last 5 years only) is 18, reflecting more recent output. A colleague in the same department, a mathematical statistician with 60 papers over 20 years, has an h index of 22 (lower, partly reflecting the smaller citation norms in mathematics) but has authored two papers with over 500 citations each, which the h index does not fully capture.