Measurement Evaluation Of Scientific Research Papers By Nobel Prize Winners In Physiology And Medicine

PurposeBased on the scientific papers of 55 Nobel Prize winners in physiology and medicine from 1981 to 2005, this study is to explore the temporal distribution and population distribution of scientific papers. The problems in the current research evaluation and their possible causes are revealed by comparative analysis during the evaluation results of multiple quantitative methods and one qualitative evaluation method(focus group discussion), so as to provide scientific basis and theoretical support for scientists'growth and success and the realization of scientific management.ObjectsThe general information of the scientific output of 55 Nobel Prize winners between 1981 and 2005 in physiology and medicine are the basis of this study. Sources of InformationThe original data are collected and synthesized by entrusting the Information Center of The Chinese Academy of Sciences.Method1. Retrieval of literature including both paper and electronic publications2. Rank—Frequency Distribution Research3. curvilinear regression and nonlinear iterative regression4. nonparameters test and analysis5. principal component analysis6. h-index method7. Analytical Hierarchy Process ( AHP)8. focus group discussionResults1. On the whole, citation frequency of world-class scientists'papers increase year by year in the 10 years after publication. Cited 225 ~ 500 times'papers reach the citation peak when they have been published for 15 to 20 years, and the cited more than 500 times reach later to 16 ~ 30 years.2. Rank—Frequency Distribution Research has also applicability in microcosmic field. Results show that the number index fits to the negative logarithm function model distribution (the independent publications are not included), and qualitative index fits to the negative exponential function model distribution. The magnitude of concentrations varies from 0.308 to 0.637, which deviates from the Price's law (0.50).3. The internal gradation instrument in the science community, the higher the measurement criteria,the greater the discrimination will be. Differentiation index of scientific papers in evaluation: SCI < group of top journals (like Cell, Nature and Science) < single top journals (like Cell, Nature and Science).4. By different quantitative research methods, the evaluation results are similar,which different to the results of qualitative analysis.Spearman relevant analysis between the result of qualitative analysis and these quantitative research methods carried out respectively, there is no correlation between the results of qualitative analysis and principal component analysis or h index(p>0.05), and a weak correlation between the results of qualitative analysis and analytic hierarchy process(r=0.38, P=0.004).Conclusion1. Journal impact factor with span of 2 years has not been well reflecting the influence diffusion law of scientific papers in biomedical field,and the span should be lengthen.2. The internal delamination in the scientific community is intermittent rather than continuous,in which scientists'further promotion shouldn't depend on the cumulative effect but on the high-quality performence.3. The Scientometrics has not been a mature discipline, for there is not a clear definition of research output "quality" and short of direct and effective research measurement indexes.4. Now scientific papers'evaluation is mechanically treated as a text assessment. The peer review relies a text assessment more and more, but not does the question assessment. The real purpose of scientific papers'evaluation is neglected. Then seeking and grasping the nodes between quantitative and qualitative analysis are the motivation for the innovation of science and technology evaluation system and responsibility of decision makers in science and technology management.Enlightenment1. Scientific research especially basic research should be based on long-term theoretical accumulation and basical work accumulation, and multiple query and thinking .It'll take a long time to vertify and identify a innovation achievement.2. It is necessary to scientific research to seek survival with quality rather than quantity.New points of view in this study1. Frstly by using the research thinking of distribution coming from epidemiology and describing of the time distribution and the population distribution of scientific papers, the sort of the discrimination of science community's internal gradation tools, the discontinuity of the scientific community and in which scientists'transition feature are obtained.2. By comparison between the results of quantitative and qualitative evaluation, the reasons that scientometrics has not been a mature discipline-- lacking of clear definition of research output "quality" and direct and effective research measurement index are discovered.Conceptions for making further investigations1. The quantitative and qualitative research on the outcome of other scientists should be carried out,so as to test the universality of the output laws of Nobel Prize Winners in Physiology and Medicine.2. To make sure the relationship between the times of cititation and time and to select the optimum time span of journal impact factor in biomedical field, more work should be done to collect more data and materials and to make a further study about scientific papers in biomedical field by birth cohort analysis.