Does Quantity Make a Difference?

Do highly productive researchers have significantly higher probability to produce top cited papers? Or does the increased productivity in science only result in a sea of irrelevant papers as a perverse effect of competition and the increased use of indicators for research evaluation and accountability focus? We use a Swedish author disambiguated dataset consisting of 48,000 researchers and their WoS-listed publications during the period of 2008-2011 with citations until 2014 to investigate the relation between productivity and production of highly cited papers. As the analysis shows, quantity does make a difference. Conference Topic Indicators; Science policy; Research assessment Introduction One astonishing feature of the scientific enterprise is the role of a few extremely prolific researchers (Price, 1963). Thomson Reuters call them Highly Cited Researchers and they are listed and recognized per area. Based on another dataset, Scopus publications, Klavans & Boyack (2015) call them “superstars” and use them for large-scale studies of publication behaviour, thereby showing that superstars publishes less in isolated areas (retrieved using a clustering procedure), in dying areas, or in areas without an inherent dynamics. Highly productive and cited researchers tend to look for the new opportunities. Obviously, the highly productive researchers have to be taken into consideration for many reasons, both for science policy and for scholarly understanding of how the science system works. Within bibliometrics there is a discussion on how to measure and to identify the superstars. Many current papers discuss the correlation between the various indicators developed for performance measurement. One of the stable outcomes is that there is a high correlation between the numbers of papers a researcher has published and the number of citations received (Bosquet & Combes, 2013). From that perspective, both indicators tend to measure the same attribute of researchers, as is actually materialized in the introduction of the H-index (Hirsch, 2005). Parallel, the discussion about impact has shifted from counting (field normalized) numbers of citations to more qualified types of citations and publications. As the progress of science rests on the huge amount of effort and publications, the number of real discoveries and path breaking new ideas is rather small. This has led to a different focus. Instead of counting publications and citations, the decisive difference is whether a researcher contributes to the small set of very highly cited papers. Different thresholds are deployed, from the top 1% or 10% of the highly cited papers or with the CCS method proposed by Schubert & Glänzel (1988). Only when reaching into these select set of papers that qualifies for citations above the x% level one can be considered as really having distinctive result that contributes to scientific progress. Increasingly, performance measures take this selectivity into account, and when calculating overall productivity and impact figures for researchers, papers (productivity) and citations (impact) are weighted differently depending on the impact percentile the paper belongs to (Sandström & Wold, 2015). Of course, the question now comes up what a good publication strategy is – given this way of performance evaluation. Is publishing a lot the best way – or does that generally lead to normal