The databases mainly offer information on the basis of citation analysis of journals, subject areas and countries and can be used, for example, to check the relative significance of journals within a subject category or - as an author - to select journals for future publications or to evaluate existing publications.
In the Journal Citation Reports Science Edition database of the American company Thomson, the article and citation data of more than 8000 mainly English-language journals from the fields of medicine, technology and science are evaluated each year. There is also the JCR Social Sciences Edition. The data is published online each year in July and is based on the evaluations of the previous year (JCR year).
The journal impact factor indicates how frequently the "citable" articles of a certain journal were cited on average in the two years before the JCR year. An impact factor of 1.0 means that, on average, each citable article of the previous two years was cited once. Citing articles may be from the same journal.
Example for journal X in JCR year 2006:
|IF =||Number of citations from 2006 to journal X in the 2 previous years|
|Number of citable articles of journal X in 2004 and 2005|
Note: "Citable articles" in the denominator are usually research articles and reviews. In the numerator, on the other hand, the citations of all document types of JCR journals are also counted amongst the non-citable document types, such as editorials, letters, news items and meeting abstracts.
This value was introduced with JCR 2007 - the time window for citable articles has been extended from 2 to 5 years.
The "aggregation" groups journals according to subject categories and is calculated in the same way as the journal impact factor, but takes into account all journals of a subject category. The median applies to all journal impact factors of the subject category.
In order to properly assess the value of impact factors, you have to consider a number of conditions and weaknesses:
Most of the issues mentioned here are taken from the article by Dong; Loh; Mondry: The "impact factor" revisited (published on 05.12.2005). Other issues are listed there.
For more detailed information, see AWMF proposals for the use of the "impact factor" (in German)
Two new evaluation criteria were developed at the University of Washington on the basis of JCR data, to rank JCR journals according to their influence or importance. The values are described at eigenfactor.org, for legal reasons they are accesible from december on for the previous year - and they are accesible starting with JCR 2007 in JCR.
Among other things the calculation considers the prestige of a journal. There is a 5 year window for the analysis of the citations. Self citations are filtered out.
The Article Influence score is a value for the average influence of articles of a journal and is therefore comparable with the journal impact factor. It is calculated with the Eigenfactor score divided by the normalized part of the articles of a journal. The median AI score is 1.
The Eigenfactor score is a value for the influence of a journal on the scientific community. It should be a measure for the time slice a user spends reading a journal. The total value of all Eigenfactor scores of the JCR journals is 100.
Other indicators are calculated in the JCR both for individual journals and for individual subject categories. Some of these include:
The "journal immediacy index" tells you how often articles published in a specialist journal are cited on average within the same year. It is calculated by dividing the number of citations in a particular year by the number of articles of the same year. Citing articles may be from the same journal.
The "aggregate immediacy index" is the value for one subject category.
Cited half-life of a journal: Half of the citations to a particular journal refer to articles that were published within the year value calculated (counted backwards from the JCR year).
For example: Journal X has a value of 7.0 years in the JCR year 2006. This means that 50% of all citations, which refer to journal X, were between 2000 and 2006 (inclusively).
The "aggregate cited half-life" is the value for one subject category.
The higher the value, the greater the average half-life of an article. Less modern or less fast-moving subject categories, in which the citation maximum is only reached after more than 2 years can, in some situations, compensate for a lower impact factor with a higher half-life by multiplying the two values, see Description of the impact factor in Wikipedia (accessed on 12.02.2016).
Citing half-life of a journal: Half of the citations in a journal refer to articles that were published within the year value calculated (counted backwards from the JCR year).
For example: Journal X has a value of 7.0 years in the JCR year 2006. This means that 50% of all articles, which were cited from articles of journal X, were published between 2000 and 2006 (inclusively).
The "aggregate citing half-life" is the value for one subject category.
This free database (Scimagojr.com) offers among other things the value SCImago Journal Rank (SJR, since 1999), which is comparable with the IF, but includes a consideration of the prestige of a journal. The basis is the database SCOPUS (competing product to Web of Science), which analyzes more journals than JCR/WoS. There is a 3 year window for the analysis of the citations. Self citations are filtered out. The data are updated yearly in November.
Via the Journal display you can see some values for subject areas or individual journals like the average citaton per document, the amount of uncited documents or the rate of international collaboration.
With the Country Rank you can analyze separate countries and 8 world regions - the values are e.g. the amount of publications, average citations per publication, self citations, uncited documents and the h index (number of publications (h) that have been cited at least h times).
Part of the Viz Tools is the Subject Bubble Chart tool to compare e.g. the subject categories in medicine by bubble charts according to different criteria.
This freely accessible online service is connecting a Top-20-Ranking of english language journals in each subject category or subcategory with links to the cited articles that are responsive for the ranking position.
The data collection will be updated each year in July and is based on articles that were indexed in Google Scholar in the last 5 completed years. There are some coverage notes.
The data is ordered by the h5-index. (For e.g. 2013 applies: "h5-index is the h-index for articles published in the last 5 complete years. It is the largest number h such that h articles published in 2008-2012 have at least h citations each".) If one clicks on the h5-index-value you will see the h articles of each journal, ordered by the number of citing articles (Cited by). Now one can navigate directly to the articles on the publisher's web page, or one can click on the Cited by link for these citations.
A comparable JCR ranking would be the 5-Year Impact Factor.
Note for article access via Metrics: Entitled users have only then access to licensed full texts from outside the faculty network if they use Metrics via the Google Scholar link on our home page or the Google Scholar page.
ESI offers e.g. citation rates (Citations per Paper) according to persons, institutions, countries and journals - always restrictable via a rough subject classification (Research Fields). Highly Cited Papers (last 10 years) and Hot Papers (last 2 years) are selectable with the same criteria.
Via the menu link Field Baselines, the average citation rates and percentiles according to separate years and the total citation rates are shown - always according to subject areas.
Research Fronts group (according to subject areas) often cited publications (Core Papers) to special topics.
On the article level there is always a link to Web of Science.