Beware of how you are measured: the h-index produces biased comparisons of scientific productivity across specialties
Stephen Haines, Truong Do, Caterine Miller, Walter Low
Introduction: The h-index is a dominant measure of academic productivity. While the h-indices of scientists publishing in the same field may be appropriately compared, the validity of comparisons across scientific disciplines has been questioned. Radicci has proposed a modification of the h-index to address this issue.
Objective: We evaluated the validity of h-index comparisons across specialties of medicine by comparing the Radicci index to the h-index for 22 medical specialties.
Methods: We curated a database of all articles published from 2002-2015 from the three highest impact medical journals: New England Journal of Medicine, the Lancet and the Journal of American Medical Association. Articles were categorized into medical sub-specialties and the respective h-index and Radicchi index of each specialty were calculated using the Web of Science publication and citation databases and classification of specialty.
Results: The Radicchi Index eliminated variability associated with publication and citation opportunity between different specialties when compared to the h-index. Using the h-index, neurosurgery tied for a rank of 15 of 22 medical specialties. Using the Radicci index neurosurgery tied for a rank of 7.
Conclusions: Comparisons of academic productivity across specialties can be used by academic administrators to assess departmental productivity and inform the allocation of resources. In this scenario, departments of neurosurgery are unfairly disadvantaged. The Radicci index, by taking publication and citation opportunity into account, produces less biased comparisons. Work by others suggests that significant differences in publication and citation opportunity exist among the subspecialties of neurosurgery. Similar concerns about h-index comparisons within neurosurgery must be raised.