Econbase downloads and the ranking of Australian university economics research: a comparative study

In recent years the policy focus of the Australian government has been towards a greater concentration of the limited research resources. This is most evident in the view that it is incorrect for the research base of a given university to be broadly based, but instead should be focused into a narrower set of areas of research strength. In this regard Kemp (1 999, pp. 10) in the formulation of policy around research and research training states, “The current structure of incentives encourages all universities to be comprehensive research institutions, rewarding them for the breadth of research they undertake. . . . If Australia is to be competitive in the global knowledge economy, we must be at world standard in every field in which we are research active. This requires institutions to determine their own research strengths and concentrate available resources to create a critical mass of internationally reputable expertise.” In more recent times Nelson (2002, pp. 25) in the context of setting up a broad review of higher education states, “However the one size fits all incentives have encouraged uniformity rather than diversification. Institutions have largely confined themselves to a predictable spread of teaching, research and research training at the expense of quality. ... Some have argued that there may be a need for shift the policy thinking from unplanned to planned differentiation, for example, the Government could designate and resource a smaller number of institutions, or a few distributed networks across a range of fields, for world class research intensity.” At a broad level the measures reported in Marginson (1999) and Nelson (2002) demonstrate that research intensity is concentrated into a small set of Australian universities. The issue is the usefulness of this information at a specific discipline level. In other words do the broad trends for research as a whole hold for a specific discipline area across a range of measures. This is particularly critical for the identification of appropriate benchmarks at a disciplinary level. The purpose of the present paper is to investigate the diversity of research strength for Australian universities in the broad discipline areas of economics, econometrics, finance and accounting. These represent the broad set of discipline areas covered in the journals, which are published by Elsevier Science Ltd. and listed on the Econbase website (http://www.elsevier.com/inca/homeDage/sae/e~onworl~menuh.t m). Since 2000 downloads of full text papers has been possible as part of the Science Direct Service. In addition for each of the journals listed on the Econbase website, the top 10 downloaded papers from the journal and the number of downloads are listed on the website. This information can be used to construct ranking lists for Australian universities on the basis of downloads for each of 2000 and 200 1. There are a number of limitations in using Econbase downloads to construct ranking lists for Australian Universities. First the set of journals on Econbase includes only two of the sixteen group 1 journals identified by Towe and Wright (1995). Thus to the extent that some universities focus their outputs predominantly in the group 1 journals they will not be captured in these measures. Second the set of journals on Econbase does not include the main Australian journals such as Economic Record, Economic Papers, Australian Economic Papers, Accounting and Finance, Australian Journal of Management. It is fair to assume that a significant portion of Australian research is focused on Australian policy issues and published in these Australian journals. This research activity is not captured in the Econbase download measures. There are a number of studies which construct worldwide ranking lists of economics research. In an extensive study Coupe (2000) uses 11 different ranking methodologies for publications and 3 different ranking methodologies for citations. For publications, Coupe (2000) finds only 2 Australian universities (ANU, UNSW) in his top 100 and a further 3 Australian universities (Melbourne, Monash, Sydney) ranked between 100 and 200. For citations Coupe (2000) finds 1 Australian university (ANU) in the top 100 and a further 4 Australian universities (UNSW, Monash, Melbourne, UWA) ranked between 100 and 200. In another study Kalaitzidakis, Mamuneas and Stegnos (2001) create a world ranking list based upon a set of top journals, where the journals are ranked by citation impact. Kalaitzidakis, Mamuneas and Stegnos (2001) rank 2 Australian universities in the top 100 (ANU, UNSW) and a further 4 Australian universities (Melbourne, Monash, UWA, Sydney) are ranked between 100 and 200. There is also a recent study by Garcia-Castrillo, Montanes and Sanz-Gracia (2002) that produces a worldwide ranking of economics research. This study is based on publications in 55 journals over the period 1992 to 1997. This study finds that Australia ranks 6*h in terms of countries which produce economics research in producing 1.169% of publications in these 55 journals. This study produces a rank list of the top 1000 organisations in the world, and there are 22 Australian organisations in the top 1000. Consistent with the findings of Coupe (2000) and Kalaitzidakis, Mamuneas and Stegnos (2001) there are only 2 Australian universities (ANU, UNSW) in the top 100 and a Wher 3 Australian universities (Monash, Sydney, Melbourne) ranked between 100 and 200. There are very few Australian universities at the top of these international ranking lists. Therefore the present paper focuses its attention on the local Australian university ranking lists provided by Towe and Wright (1995), Towe (1996) and Sinha and Macri (2002), and the Asian ranking list provided by Chan, Chen and Steiner (2001). These more local ranking lists contain a much broader set of Australian universities. The ranking lists constructed on the basis of downloads can then be compared to ranking lists which have been constructed on the basis of publication outputs by Towe and Wright (1995), Towe (1996), Chan, Chen and Steiner (2001) and Sinha and Macri (2002). In addition ranking lists can also be constructed on the basis of performance in Australian Research Council (ARC) grants as a basis for further comparison. The ARC runs the national competitive grants program and details are available on its website ( http://www.arc.gov.au/ncm ). The plan of this paper is as follows. In section 2 the ranking lists based on downloads are compared to ranking lists based on publication outputs and grant outcomes. Section 3 makes some concluding remarks about the interface of these ranking lists with the broad question around concentration of research activity