Scientometrics, VoL 20. No. 2 (1991) 359-362 THE BIBLIOMETRIC ASSESSMENT OF UK SCIENTIFIC PERFORMANCE - SOME COMMENTS ON MARTIN'S "REPLY" T. BRAUN,* W. G~ZEL, A. SCHUBERT Information Science and ScientometricsResearch Unit (ISSRU) Library of the HungarianAcademy of Sciences P.O.Box 7, H-1361Budapest (Hungary) *also with the Institute of lnorganic and Analytical Chemistry, L. Ei~tv6s University, P.O.Box 123, H-1443Budapest (Hungary) (Received June 27, 1990) No new arguments or evidence that undermine our conviction that available seientometric measures do not indicate a statistically significant "decline" of British science in the first half of the eighties have been found in Martin's reply. Martin's meticulous scrutinyI actually confu'med the main message of our "Flash "2, i.e., that extremely inconsistent conclusions can be drawn from the very same set of data using formally correct arithmetical manipulations. We restrain, therefore, enumerating any more arguments in the continuing debate on the decline of British science until a considerable amount of new data becomes available. 3 In what follows we are commenting on some of the methodological questions alluded to in Martin's "Reply ~. 1. Our critic pinpoints as one of our supposed "flaws" that we did not take into account that "an increase in the absolute number of papers produced by a single country [...] may or may not correspond to a genuine increase in scientific output." Although it is not quite clear what is to be understood under the term "genuine increase", this assertion seems to be practically another version of what was formulated in our "Flash" as "any attempt to measure national performances by total [publication or citation] counts proved to be uncontrollably corrupted". 2. In our "Flash" we naively reported the range and the average of the "percentage share" indicators and of their "annual mean relative change". Never in our wildest nightmares would we have thought that one of these innocuous figures (+0.93%) would be practically considered the "final score" of our whole exercise and would be attacked accordingly. We rather intended to put the emphasis on the ranges of the Scientometrics 20 (1991) Elsevier, Amsterdam-Oxford-New York- Tokyo Akad$miai Kiad6