Quantitative measures of communication in science: a critical review.

TLDR: It is the contention that the tendency to characterize 'science' in terms of this formal process, and to employ methods designed to analyze the formal literature, "obscures key features" of scientific research.

Abstract: ed, usually colloquial, frequently incomplete, and often vague. The communicator here is not seeking to report a finished scientific work. He often knows, in fact, that the person with whom he is communicating needs only a minimal communication of an idea to understand fully its meaning and importance for their common subject of research. The recipient embodies integrated knowledge; therefore, the message need not, in itself, be integrated. 49 One is tempted to say that formal communication in science is 'the tip of the iceberg', were it not for two facts: (a) the 'tip' is very large, extensive and important; and (b) there is every indication that the 'tip' is radically dijjerent in kind from what is 'below the waterline'. (Perhaps 'the soft underbelly of science' might be a more appropriate metaphor I) The difference is signalled in Garvey and Griffith's very next sentence: "Such loose communication cannot, of course, be tolerated in the journal literature of science where a more universal audience exists."50 Formal 'communications' in science represent a process of assimilation: of "separating scientific fact from conjecture", 51 and of "the transformation of research findings into scientific knowledge". 53 The object of a formal research paper is to persuade and convince the appropriate audience that the results presented should be accepted as valid knowledge: the form and style of argumentation is determined by that institutionalized goal, and the paper is assembled, and its list of references compiled, with that end in view.P It is an 'after the event' view, re-ordered and re-presented in accordance with established canons of proper practice; referees and editors ensure that certain decencies (including adequate referencing) are observed. It is not at all surprizing that Garvey and Griffith, comparing final published papers with their antecedent technical reports, found that "the corresponding journal articles were typically better written and better related to other work in the same subject areas". 54 'Communication', as it is usually understood, appears to be a relatively minor role of formal 'communications': While 'current' journal reading is relevant to the dissemination of research findings, it seems to be a minor portion of the use to which journals are put. For example, it amounted to only about one-third of the journal reading of one group ofextremely active psychologists.55 ... it was quite clear in 1962 that the journal article in psychology. was no longer the medium for disseminating current scientific findings to researchers active on the research front. 56 . What, then, is its 'use' ? Since most citations are given to research papers, how can these be assumed to reflect an influence? Gaston claims that "The QUANTITATIVE MEASURES OF COMMUNICATION 115 scientific community must wait until every aspect of the research is published before it can use the results to guide new research or correct an earlier perspective".57 Is this true? As we have seen, Cole seems to equate 'use' with 'citation' -that is, with the process of public 'assimilation' in the formal literature. And even Garvey and Griffith seem to define 'science' so that the formal 'tip' characterizes the whole: they note that, on average, some two or three years after publication, an article ... is cited by another author. Here the scientific information in the article is built upon, evaluated in the light of new information, and linked to new information which has been generated since its publication ... analysis, evaluation and synthesis have been proceeding ever since the work first appeared in the journal literature, in a process so central to science as to be virtually indistinguishable from it. 58 It is my contention, as I said earlier, that the tendency to characterize 'science' in terms of this formal process, and to employ methods designed to analyze the formal literature, "obscures key features" of scientific research. I t diverts attention from "the soft underbelly" : worse, it proposes that the informal should be understood in terms of insights gained by the study of the formal. I submit that this is to reverse the priorities of explanatory logic. Explanations of scientists' behaviour in the informal domain should surely be extended so as to include within their scope the formal aspects-including the relatively trivial behaviour of adding citations to papers. But, quite apart from this 'logical' point, it is simply my judgment that illumination is more likely to accrue from this contrary strategy. To claim that co-citation methods can locate actively interacting groups and specialty boundaries is not to claim that such features can only, or most economically, be discovered in this way-nor that these methods uncover (or point to) sufficient, or sufficiently accurate, detailed andrelevant, information on the groups and specialties. To refer back to the 'gas laws' analogy, and the claim that scholars must await the establishment ofnumerical patterns before they can know what 'problems' they have to solve, my response is to suggest that the everyday, detailed behaviour of scientists in the conduct of their research provides an abundance of problems of much more obvious importance than any correlations contained in a computer printout. Whenever a scientist (or a research group) decides to develop a new technique, or to pursue a fresh and unexpected phenomenon, or to adopt a perhaps unfashionable theoretical approach, there is a (sociological or historical) problem: each decision brings together 'cognitive' (intellectual, technical, cultural) and 'social and historical' factors: one task of the historian (or sociologist) of science is to explicate such decisions, and to explore the 'grounding' of their rationality. 59 In relation to this task (or,