Quantifying journals

·

Impact factor

https://en.wikipedia.org/wiki/Impact_factor

  • Average (mean) number of citations in the last 12 months of articles published in the previous 2 years

impact = c(year 0) / (n(year –1) + n(year –2))

e.g. 6 = 600 / 100

  • Publish more articles = lower impact factor => selectivity
  • Doesn’t matter which articles get the citations (could be just one)
  • If one article out of 100 published has 600 citations, and the rest have none, impact factor = 6

Pros: publishing more, low cited articles reduces the impact factor

Cons: Distribution can be highly skewed, affected by a few highly cited papers

h-index

https://en.wikipedia.org/wiki/H-index

h-index = max(n(articles with n citations))

  • Can publish 10000 articles with no citations, h will stay the same
  • If 60 articles out of 60000 published have 60 citations, and the rest have none, h-index = 60

Pros: less affected by a few highly cited papers (n represents how many of those there are)

Cons: can publish many low cited articles without reducing the score

Eigenfactor

https://en.wikipedia.org/wiki/Eigenfactor

eigenfactor = n citations, weighted by citing journal rank

  • Can publish 10000 articles with no citations, e will stay the same
  • Doesn’t matter which articles get the citations (could be just one)
  • If 1 article out of 60000 published has 60 citations from journals with highly-cited articles, and the rest have none, e = 60