Wednesday, 17 February 2016

H- and G-indexes... and Harzing's index

Like all human endeavour, we have the leaders in any field and the also-ran's. We humans seem to like differentiating things and deciding what equals things of higher or lower status and value: whether that be cars, clothing, music, art or writing.

So, you can understand why evaluating academic journal quality is a bit fraught when you are comparing different disciplines, longevity, universities, funding regimes, countries of origin, and  philosophical bases.

Google Scholar makes it extremely simple: it looks solely at how many times a piece of work has been cited by others. However, Google Scholar can't, with a citation, measure whether an article has been cited to refute what it says or to uphold it.

Lots of people have had a crack at more complex models: like the h-index and g-index.

The h-index was developed by Jorge Hirsch in 2005 (so quite recent), as a measure of writer's productivity and the number of citations their publications get over time. It will start low at the beginning of an academic's career, and hopefully increase over time as they get better known in their field (ie, "h of [their] N papers that have been cited at least h times each, while the rest of the N papers have less than h citations each"). Then in 2006, Leo Egghe created the g-index to measure productivity based on the writer's publication record which has a more complicated mathematical formula (something like the greatest number that the top G articles received altogether, at least G square citations). Both these indexes are used as well as the journal rankings to say whether a particular academic has cut through or not.

Then there's SCImago, the World of Knowledge database (thus Thomson Reuters Web of Science), the Aston Business School, the Vienna University of Economics and Business Administration (WU Wien), the Danish Ministry, HEC (Hautes Études Commerciales de Paris), the University of Queensland Excellence in Research Australia, AERES (Agence d’évaluation de la recherche et de l’enseignement supérieur), the Cranfield University School of Management, the Erasmus Research Institute of Management, the Australian Business Deans Council, FNEG (Foundation National pour l’Enseignement de la Gestion des Entreprises), The Association of Business Schools (led by Bristol and Harvey Morris).

A wonderful professor, Anne-Wil Harzing, does a table each year rating all the business management journals, which she publishes as open access on her website (here).

The Harzing journal ranking also includes the RAE ranking evaluated by Mingers and Harzing (2007), which considers a whole heap of data to come up with a summary solution, the RAE Level Descriptor:
  • 4: Quality that is world leading in terms of originality, significance, and rigour.
  • 3: Quality that is internationally excellent in terms of originality, significance, and rigour but which nonetheless falls short of the highest standards of excellence.
  • 2: Quality that is recognised internationally in terms of originality, significance, and rigour.
  • 1: Quality that is recognised nationally in terms of originality, significance, and rigour.
  • Unclassified: Quality that falls below the standard of nationally recognised work. Or work which does not meet the published definition of research for the purposes of the assessment.
I like this latter measure. It seems eminently sensible to me to use one that looks at citations, h-index, g-index, university standings, and the publications themselves.


  • Harzing, Anne-Wil (2015). Journal Quality List. Retrieved 7 January 2015 from
  • HLWiki Canada (2015). Author impact metrics. Retrieved 7 January 2015 from 
  • Mingers, John & Harzing, Anne-Wil. (2007). Ranking journals in business and management: a statistical analysis of the Harzing data set. European Journal of Information Systems, August 2007, Volume 16, issue 4 (pp. 303-316).
  • Research Gate (21 December 2012). What is the difference between H-index, i10-index, and G-index? Retrieved 7 January 2015 from

No comments :

Post a Comment