Page 130 - Academic and Research Book_WEB

Author Level Metrics

The citation based metrics for journals can easily be extended to the case of authors which have
implications for their productivity as well as measure their impact on the scientific community.
We have already emphasized the importance of citations, which can easily be extended to deduce
contributions of authors at their individual or collective levels. This not only provides the first
glimpse of scientist contributions but also lays the foundation of collaborations between scientists
and institutions. The author-level metrics also provide a tool for assessing the research contributions,
which play an important role in their career progressions. However, it needs to be emphasized that
these cannot be the sole criteria as stressed by Eugene Garfield himself, it has to be supplemented by
peer or other supporting review mechanisms.

h-index

h-index is perhaps the most widely-known author level index and is very extensively used as a proxy
for author’s academic achievements in the research domain .This index quantifies both the scientific
productivity and the apparent impact created by a scientist in the scientific world. The index is based
on the set of the scientist’s most cited papers and the number of citations that these have received
in other people’s publications as well as through self-citations. The index can also be applied to the
productivity and impact of a group of scientists, such as a department, university, or country.
In 2005, Jorge E. Hirsch, a physicist at University of California at San Diego, USA, first suggested this
index and it is sometimes called the Hirsch index or Hirsch number. (Hirsch, 2005). h-index is calculated
by number (h) of an author’s articles, which have been cited at least same number (h) of times. For
example, if an author has published 20 papers or more and 20 of these have each been cited 20 times
or more, then the author will have an h-index of 20 or h = 20.
The h-index seems to give better qualitative information than the total number of research
publications as well as the total number of citations received. Merely knowing the number of
publications does not reveal how well these articles have been received by other researchers.
Similarly, the total number of citations can be inordinately influenced by a small number of
highly-cited few articles, in which case it may not be a true reflection of one’s productivity, or
a large number of poorly-cited papers, in which case the quality of work may not be up to the
mark.

The main advantage of the h-index is that it is neither skewed upwards by a small number of highly-
cited papers nor skewed downwards by a long tail of poorly-cited work. A high h-index indicates that
the research work is receiving adequate attention in the literature as well as shows the consistency and
the contemporary nature of the research of a particular author. From July 2011 onwards, Google has
provided an automatically calculated h-index within their own Google Scholar profile.

The Limitations

Before using h-index as the marker for research quality and output for an author, several observations
have to be kept mind in its interpretation:

Results can be inconsistent: Although the basic calculation of the h-index is clearly defined, it can be
calculated using different databases or time-frames, giving different results. Normally, the larger the
database, the higher the h-index calculated from it. Therefore, an h-index taken from Google Scholar
will nearly always be higher than one from Web of Science, Scopus, etc.

ACADEMIC INTEGRITY AND RESEARCH QUALITY | 119