Citation counts often are used to measure scientific accomplishment. It is very difficult, however, to compute accurate citation counts in research where one has a list of scientists but not their complete bibliographies. At the same time, procedures are available that permit informed judgments about whether given scientists were cited at all in given years. The possibility of such judgments suggested that the number of years in which scientists were cited might be used as an alternative measure of scientific accomplishment. This possibility was explored in two studies, one based on 2,713 population scientists and the other on 135 articles published in the journalFertility and Sterility. Years cited was easy to compute, and had good descriptive statistics, satisfactory generalizability coefficients, high correlations with total citation counts, and distributions little influenced by outliers. These results supported the appropriateness of the years cited measure.
Studies of journal citation patterns suggest that specialty areas within disciplines may be the most appro priate structural units for understanding the social organization of science. Citation studies necessarily are limited to scientists who publish, however, and studies of all members of particular disciplines would provide more general specialty structure data. Accordingly, this research applied factor analytic procedures previously used in studies of the structure of specialization among psychologists to all members of the Population Association of America. Four principal components derived from the self-designated specialties of these population scientist were rotated to a final solution by the varimax procedure and were interpreted as measuring, respectively,Social Emphasis, Geographic Emphasis, Formal Emphasis, andEpidemiological Emphasis. These results partially confirm the distinction sometimes made by population scientists between social demography and formal demography, but suggest this typology is incomplete. The results also illustrate techniques that could provide a useful alternative to citation analysis for researchers studying specialty structures in other disciplines.
Authors:C. Angell, J. Green, K. Ito, P. Lucas, and B. Richards
In this paper we introduce two key notions related to understanding the glassy state problem. One is the notion of the excitation profile for an amorphous system, and the other is the notion of the simple glassformer. The attributes of the latter may be used, in quite different ways, to calculate and characterize the former. The excitation profile itself directly reflects the combined phonon/configuron density of states, which in turn determines the liquid fragility. In effect, we are examining the equivalent, for liquids, of the low temperature Einstein-Debye regime for solids though, in the liquid heat capacity case, there is no equivalent of the Dulong/Pettit classical limit for solids.To quantify these notions we apply simple calorimetric methods in a novel manner. First we use DTA techniques to define some glass-forming systems that are molecularly simpler than any described before, including cases which are 80 mol% CS2, or 100% S2Cl2. We then use the same data to obtain the fragility of these simple systems by a new approach, the 'reduced glass transition width' method. This method will be justified using data on a wider variety of well characterized glassformers, for which the unambiguous F1/2 fragility measures are available. We also describe a new DTA method for obtaining F1/2 fragilities in a single scan. We draw surprising conclusions about the fragility of the simplest molecular glassformers, the mixed LJ glasses, which have been much studied by molecular dynamics computer simulation.These ideas are then applied to a different kind of simple glass — one whose thermodynamics is dominated by breaking and making of covalent bonds — for which case the excitation profile can be straight-forwardly modeled. Comparisons with the profile obtained from computer studies of the molecularly simple glasses are made, and the differences in profiles implied for strong vs. fragile systems are discussed. The origin of fragility in the relation between the vibrational and configurational densities of states is discussed, and the conditions under which high fragility can convert to a first order liquid-liquid transition, is outlined.