• Identifies items that have cited a specific work, or author, or set of works
• Helps you to find later research, based on earlier work of known value
• Complements traditional keyword or subject searches.

Scopus is the multidisciplinary scientific citation database of widest title coverage. 

Web of Science is the multidisciplinary scientific citation database of widest date coverage.
Science Citation Index — is included in Web of Science.

You may create alerts to be notified when a specific work or author is subsequently cited. For help with these multidisciplinary databases, contact Michael Parkinson (Subject Librarian: Mathematics, Statistics, & Physics); but for more subject-specific citation resources, ask your Subject Librarian for help.
[Researchers at other Faculties should consult their own Subject Librarians]

You may also consult the guide Measuring and Evaluating Science Research (49 kb, 2 pages), prepared by the science subject librarians.

Impact Factor  (IF)  or  Journal Impact Factor

Published in the Journal Citation Reports (JCR), impact factors are intended to measure how often, on average, authors cite moderately recent articles from journals.  The citation data is from the same source as Web of Science.  For 2009, which is the latest version of JCR and was relased June 2010:

Article Influence

[Also published in Journal Citation Reports] The article influence of a journal expresses the average influence, per article, of the papers in that journal. It is like the 5-year Impact Factor, except the citations are weighted to reflect the prestige of the citing journals; the results are normalised (i.e., the mean article has an article influence of 1.00). See http://www.eigenfactor.org for more details.

Eigenfactor Score

[Also published in Journal Citation Reports] For a journal's impact factor, each citations in a year has equal weight; but the Eigenfactor gives more weight to those citations made in more influential journals (see the website http://www.eigenfactor.org for details of the algorithm). All the eigenfactors are scaled so as to add up to 100 for each particular year and edition, hence the "Eigenfactor™ score of a journal is an estimate of the percentage of time that library users spend with that journal."  

SCImago Journal Rank (SJR)

Like Eigenfactor.com, the SJR attaches weightings to the various citations of a journal according to the prestige of the journal in which those citations appear.  For details of the SJR algorithm, as well as the rankings, see the website http://www.scimagojr.com/  Because SJR uses citation data from Scopus, it covers more scientific journals than JCR.

WARNING:  It is misleading to compare journals based solely on their citation measures.  Citation patterns vary across different disciplines and subdisciplines.  Even within a single subdiscipline, journals of identical "quality" will have differing impact factors (or article influence, etc) because of differing content: the ratio of "articles" (compared with letters, editorials, etc) and the conceptual nature of those various articles (review articles, methods papers, etc).  Although the impact factor is a comparatively transparent and verifiable metric, it is still subject to manipulation and innate inequities.

Here is an example of a researcher with an h-index of 21      

h-index

IF h is the greatest integer such that a particular researcher's h most-cited papers have each been cited h times or more, THEN h is that researcher's h-index.

g-index

IF g is the greatest integer such that a particular researcher's g most-cited papers have been cited, in total, g² times or more, THEN g is that researcher's g-index
The g-index is a modification of the h-index, to correct for scientists with highly cited works.  Clearly g≥h and, in this case, the researcher has a g-index of 29 since the SUM of the citations of the top 29 papers is 863≥841=29², but the SUM of the citations of the top 30 papers is 811<900=30²

There are other suggested modifications of the h-index, intended to "compensate" for the number of co-authors, or the time-span since the author first published, or whether the author is still publishing, etc.

WARNING:  The "same" metric usually varies from database to database (eg, an h-index from Scopus or from Web of Science), because of their different journal coverage, time-periods, or citation verification.

PDFs for librarians on ... Finding more citations in Scopus (400 kb, 2 pages) and in WoS (in preparation); Creating a ResearcherID (to be updated).

Other introductory material may be found at Research Trends (published by Scopus).  For more details on the various metrics, see your Subject Librarian or the contact below.