Death comes as the end — Effects of cessation of personal influence upon rates of citation of astronomical papers

Abstract

DEATH OF COMES PERSONAL AS THE END INFLUENCE OF EFFECTS UPON ASTRONOMICAL RATES OF OF CESSATION CITATION PAPERS V. T r i m b l e Astronomy Pro#rarn, University of_Maryland, College Park, Maryland 20742, USA and Department of Physics, University of California, Irvine, California 92717, USA A n attempt is made to investigate the extent to which direct, personal influence on students, colleagues, and mentors affects the frequency with which a scientist's papers are cited by others working in his field. The method used is a n analysis of changes in citation rate between 1965 and 1984 to papers written over their entire careers by a n index group of astronomers who died between 1969 and 1982. These citation histories are compared to those of papers written over the same period by a control group of astronomers who were still activ› at the end of 1984. Only papers written during the lifetimes of the index astronomers are considered. The data show signs that the death of a scientist is followed by a few-year sympathy period , during which citation rates rise, and then by a decade or more of gradual forgetfulness, during which citation rates drop off faster t h a n those to papers written during the same period by living astronomers. The amounts of the rise and decline suggest that about 4 0 ~ of astronomical cita- tions are mediated by personal influence. 1. I N T R O D U C T I O N Most of us believe that the frequency with which a scientific paper is cited depends primarily u p o n how useful it is to other workers in the field. Most of us also suspect, however (especially when our own papers go uncited), that there are other factors, related to where, when, and by whom the paper is published. One of these factors is direct, personal influence on mentors, colle- agues, and students with whom the author communicates. R a o and Vahia [1] have gone so far as to suggest that this personal influence factor may be sufflciently i m p o r t a n t to account for the rough proportionality that exists between the n u m b e r of authors of a n astronomical paper and the n u m b e r of times it is cited [2]. Along the same lines, the fact that astronomers typically write their most-cited papers between the ages of 50 and 60 [3] has been attributed by Woltjer [4] to the peaking of personal influence (in the form of directorships, editorships, chairmanships, etc.) during that decade. The intent here is to probe the effect of personal influence on citation rates. The method is a comparison of the citation histories of papers written by 34 astronomers who died between 1969 and 1982 to the citation histories of papers written by a control group still active in astronomy at the end of 1984. The control individuals were matched as nearly as possible to the index ones in subdiscipline, gender, country of employment, and (most important) year in whi™ career began. This last is important, because average citation rates to astronomical papers peak 5 yr after publication and decline monotonically, and roughly linearly, thereafter, with a half-life near 20 yr [5]. The death dates for the sample were constrained a t o n e end by the appearance of the first cumulative issue of Science Citation Index including astronomical papers (1965--69) and, at the other end, by the time needed for the astronomical community to be aware of the death when writing papers published in 1984. The usable sample is hot very large, leading to a preliminary conclusion that investigations like this require a larger population (or a n older one!) t h a n the world's 7000 research astronomers (median age a b o u t 40) to achieve great statistical significance. Czech. J. Phys. B 36 [1986]