Citations acknowledge the work of others and also indicate to readers where information in a paper came from.
Citations to one's publications are an indication of readership, interest and usefulness (impact) of your work to others.
Many databases now provide citation statistics and compile other metrics such as the h index. Two of the most notable such databases are Scopus and Web of Science which are listed in the Databases option on the Library website.
If you have a Google Scholar profile you will also be able to view an h index.
As disciplines differ in the way journals are published (number; issues per year, format, type of documents included) it is not possible to compare metrics across disciplines and between researchers.
A metric is a system or standard of measure. Citation Counts simply total the number of times your article, book, or other published research has been cited. Citation counts may vary across databases, such as Scopus, Web of Science, and Google Scholar, as each database indexes different journals.
One is more concerned with the impact of research output; hence, a whole range of measures of impact have come into being.
The most widely used is the h index but here are others as well.
Publishing in South African journals
Note that very few South African journals are included in Scopus and Web of Science, particularly in the humanities and social sciences. Sabinet the database of South African collections of publications has never done citation indexing so will not offer you an h index.
The WOS core collection includes some 22 000 journals, books and conference proceedings. Metrics in WOS are only for those publications included in WOS or cited in WOS publications and cover journal articles only; it does not count citations from books, conference proceedings, technical papers, theses etc so they may be lower than for Google Scholar. Cited references are from the year 1900.
h indexes are only available from 1995. An h index in Scopus is only available for items indexed in Scopus. Books and conference proceedings are included in the h index calculation. The Scopus h index is likely to be lower than the h index in Google Scholar.
SciVal is a research performance assessment tool that allows analysis of the data from Scopus. It provides country, regional, institutional as well as author level data.
GS metrics only cover a fraction of scholarly publications from the last 5 years. GS covers publications since 2019 and excludes dissertations; court opinions, patents; publications with fewer than 100 articles published between 2019 and 2023; publications that received no citations to articles published between 2019 and 2023.
The main databases for finding pre-calculated h indexes are Web of Science, Scopus, SciVal and Google Scholar. UKZN subscribes to first 3.
The h-index can be regarded as a measure of the number of publications published (productivity) as well as how often they are cited (impact).
To calculate the h-index, only two pieces of information are required: the total number of papers published and the number of citations for each paper. Where these two numbers intersect is the h-index. Articles by an author will each have a different number of citations. Where the number of publications coincides with the number of citations, this is the h index. For instance, a researcher with 17 published articles has 12 that have been cited 12 times. The h index is thus 12.
Over time one expects publication output to increase and so should the h index.
The h index of an author will be different for different databases. This is because databases differ in the time period for which they calculate the h index, they include different collections of journals etc. For instance, Scopus only considers work from 1996 or later, while the Web of Science calculates an h-index using all the years that an institution has subscribed to.
One of the 'problems' with the h index is that it does not accommodate individual papers that may have a large number of citations, and the h index stays the same even when a researcher is no longer productive.
J Hirsch who developed the h index proposes that after 20 years of research, an h-index of 20 is good, 40 is outstanding, and 60 is truly exceptional.
Other metrics provided by some databases include a number of downloads, social media accesses and more.
The i10 index or score is only offered by Google Scholar and is a creation of Google Scholar. Basically the i10 score indicates the number of articles by an author that have received 10 citations. If you have a GS profile you will see your GS i10 score and h index:
| All | Since 2019 | |
|---|---|---|
| Citations | 1437 | 618 |
| h-index | 22 | 13 |
| i10-index | 44 | 20 |
This measures the global citation performance of a set of articles and accommodates highly cited papers which teh h index does not.
To calculate the G index:
multiply the total number of papers published by the total number of citations for the most cited paper. Divide this by the number of years since the first paper was published. Then take the square root of the result to get the G index (in the Google search box put your number followed by 'square root' and a calculation will be done for you).
Eg 56 papers x 12 (being the highest number of citations for any paper) = 672. Divide this by the number of years since the first paper was published eg 6 years = 112. Now take the square root of 112 = 10.583. G index is 10.
The G index is often equal to or larger than the h index. The G index will vary depending on where you source the citation data as databases are all different. So you may well have several G indexes like h indexes.
The Publish or Perish free software can be downloaded to calculate your G index per databases, or you can do it manually (which may be quicker). https://harzing.com/resources/publish-or-perish
PoP is an interface with databases, it does not maintain databases of journals or anything else. With subscription databases you may be asked to provide the API key (application programming interface) in order to calculate the G index. Follow the instructions to access the API key.
The h and G indexes will always favour quantity and will not accommodate for example the person who has a few groundbreaking papers.