Premium
The Relevance of the “h‐” and “g‐” Index to Economics in the Context of A Nation‐Wide Research Evaluation Scheme: The New Zealand Case
Author(s) -
Anderson David L.,
Tressler John
Publication year - 2013
Publication title -
economic papers: a journal of applied economics and policy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.245
H-Index - 19
eISSN - 1759-3441
pISSN - 0812-0439
DOI - 10.1111/1759-3441.12026
Subject(s) - weighting , relevance (law) , ranking (information retrieval) , context (archaeology) , index (typography) , journal ranking , scheme (mathematics) , citation , frame (networking) , actuarial science , econometrics , operations research , computer science , statistics , economics , mathematics , library science , information retrieval , political science , medicine , geography , law , mathematical analysis , telecommunications , archaeology , radiology , world wide web
The purpose of this paper is to explore the relevance of the citation‐based “h‐” and “g‐” indexes as a means for measuring research output in economics. This study is unique in that it is the first to utilise the “h‐” and “g‐”indexes in the context of a time‐limited evaluation period and to provide comprehensive coverage of all academic economists in all university‐based economics departments within a nation state. For illustration purposes, we have selected the New Zealand's Performance‐Based Research Fund ( PBRF ) as our evaluation scheme. To provide a frame of reference for “h‐” and “g‐”index‐output measures, we have also estimated research output using a number of journal‐based weighting schemes. In general, our findings suggest that “h‐” and “g‐”index scores are strongly associated with low‐powered journal ranking schemes and weakly associated with high powered journal weighting schemes. More specifically, we found the “h‐” and “g‐”indexes to suffer from a lack of differentiation: for example, 52 per cent of all participants received a score of zero under both measures, and 92 and 89 per cent received scores of two or less under “h‐” and “g‐” respectively. Overall, our findings suggest that “h‐” and “g‐”indexes should not be incorporated into a PBRF ‐like framework.