Indicators and h-index
There are many types of bibliometric indicators that measures scientific productivity or outreach. h-index has become increasingly popular as an easy and concise way to measure publication activity, but what does it mean?
Basic bibliometric indicators
- number of publications and citations
- number of publications and citations per researcher
- number of citations per publication
- number of self citations
- number of uncited articles
- impact factor (journals, Thomson Reuters)
Basic indicators can not be used for comparisons between different subject areas.
Advanced bibliometric indicators
These indicators are always normalized, the publications are compared with other publications with similar prerequisites, for example the same area of research, the same year of publication, the same document type.
- field normalized citation score - eg. CPP/FCSm and MNCS
- field normalized top publications, Top 5%, indicates the section of publications from a research team that belongs to the 5% most cited in the world.
Field normalized indicators can be used for comparisons between subject areas.
Publication patterns - sometimes displayed as maps of co-publication and -citation between researchers (see for example Citation Map i Web of Science), co-publication between subject areas, co-publication with other organizations and other countries.
h-index is a measure that takes both productivity and impact as indicated by number of citations into consideration. It can be calculated for a collection of publications, for example one researcher's production, the whole content of a journal or all publications from a research team. Index h is defined as the number of publications that have been cited h or more times.
How to calculate h-index
When you indicate an h-index, it is important to specify the database in which it was calculated, and note at what point the h-index has been calculated. The following three databases have slightly different coverage in terms of journals, and therefore can not get the exact same h-index .
Find and mark the publications that should be included in the analysis of the database you have chosen to work with.
- Web of Science: Click "Create Citation Report" in the upper right corner of the list of search results. This will take you to a new page containing citation statistics for all publications in your list, including h-index for the list.
- Scopus: In Scopus, click "View Citation Overview" in the upper left corner of your selected list.
- Google Scholar: Create an account in Google Scholar Citations to be able to calculate h-index.
h-index for comparisons
From diverse citing cultures in different research areas follows that h-index will come out differently. H-indexes for researchers in physics are generally much higher than for researchers in social sciences for example. As h-index is based on total production of publications, it will grow with time; the longer a researcher has been active, the higher her or his h-index will become. If h-index is to be used for comparison, it is necessary to take the time dimension into account, and also to compare only with other researchers in the same area of research.
The Journal Impact Factor from Thomson Reuters is much used, but there are other relevant journal ranking systems, as well as an increasing number of more or less deliberate falsifications.
Thomson Reuter’s Journal Impact Factor (JIF)
The first (and still most commonly used) bibliometric journal level indicator is the Journal Impact Factor: "In any given year, the impact factor of a journal is the average number of citations received per paper published in that journal during the two preceding years". Impact factors according to this algorithm has been calculated yearly starting from 1975 for journals that are indexed in the Journal Citation Reports database (Thomson Reuter), using data from Web of Science or its predecessors.
SCImago Journal Rank indicator (SJR)
SJR by SCImago is calculated on the articles and citations in the Scopus database, and using a slightly different algorithm than Thomson Reuter´s Journal Impact Factor. It is calculated over three years, limits self-citations and takes into account also the impact of the journals in which citations appear. The SJR and developments can be found in Scopus under Compare journals.
Source Normalized Impact per Paper (SNIP)
SNIP takes differences in citation practices into account by comparing with the total number of citations in the citing journals. The impact of a single citation is given higher value in subject areas where citations are less likely, and vice versa. It is based on the bibliographic data in Scopus and can be found in Scopus under Compare journals.
The Norwegian list
The Norwegian model is used for evaluation and allocation of research funds in Norway, and in modified versions in universities in other Nordic countries, eg. Stockholm University. It is based on a register where scientific journals are grouped in two levels. Level 1 means scientific journals and level 2 the scientific journals which are perceived as the most leading in their research area. Level 2 should represent about 20% of scientific publications. You find the lists over journals, book publishers and other publication channels at the NSD website.
False or misleading Impact Factors
During the last years, a flora of new non-transparent impact factors have been proposed and marketed by different companies, with minimal or no documentation about how they are calculated. The main idea seems to be to sell an appearence of influence and quality to journals that are new, undistinguished or in fact frauds (predatory journals). Among the suspected IF:s can be mentioned the Citefactor, the Universal Impact Factor, the Global Impact Factor, the Scientific Journal Impact Factor and IndexCopernicus, but there are many more provided by a number of companies that offer misleading metrics.
Significance of journal level indicators
Journal level indicators for the journals in which a particular researcher has published articles is sometimes used to evaluate the work of researchers or research groups. This use is controversial, even when using solid indicators. A high Journal Impact Factor, for example, can result from a few very highly cited articles, and does not guarantee the quality of each article in the journal. However, a high Thomson-Reuters' JIF or SCImago Journal Rank is a sign that the journal has a wide and productive audience, and thus might be something to consider when choosing a journal to publish in.