Journal Impact Factor (JIF)
JIF measures the average citation rate for articles from the two previous years in a particular journal. IE: the number of citations in year x received by articles published in journal y in the years x-1 and x-2 divided by the number of articles published in y in the same years. Read more about the calculation of JIF, and some critical remarks, at PhD on Track.
JIF is based on citation data from Web of Science.
The fact that an article is published in a journal with a high impact factor, is no guarantee that it will be much cited, albeit there often appears to be a connection.
In many fields it makes little sense to count citations from such a limited time window as that of journal impact factor (two years). Some journals compensate by calculating the impact factor on the basis of citation data from a longer period, eg. five years.
The impact factor can only be used to compare similar journals within the same academic field, because different disciplines have different practices for how and when to cite. As JIF reflects the average number of citations for a set of articles over time, it does not say anything about the individual article or researcher.
Like other bibliometric indicators, JIF is vulnerable to errors. It can be manipulated, and is influenced by systematic skewing of citations (The Matthew Effect). High impact does not necessarily mean high quality; negative citations will contribute to raising the impact factor.
Like all indicators based on data from Web of Science, JIF is only relevant to academic fields in which the database covers a considerable share of journals.
Bibliometric experts are unison in their scepticism towards uncritical and uninformed use of impact factor, as worded in the DORA declaration. If using impact factor as a criterium when assessing potential channels for publishing your research, do not let it be the only one.