The Research Of Science And Technology Evaluation Based On Scientific Development View

Constructing science and technology evaluation system is an important aspect to implement the scientific development view. Science and technology are primary productive forces, which can promote economic and social development, and they are important manifestation of a country's comprehensive national strength. The establishment of comprehensive evaluation index system of science and technology as well as to evaluate our various provincial-level administrative regions has important significance to guide China's science and technology development.This paper is divided into five chapters. Chapter 1 is introduction. Chapter 2 constructs science and technology evaluation index system model based on Cluster-Factor Analysis. Chapter 3 constructs science and technology comprehensive evaluation model based on the G1-maximizing deviations in portfolio empowers. Chapter 4 is empirical research on typical provincial-level administrative region of China on science and technology evaluation. Chapter 5 presents conclusions.This major work: first, under the "people-oriented, comprehensive, coordinated and sustainable development" scientific development, this paper uses R cluster and factor analysis to establish a comprehensive evaluation index system of science and technology. Second, it uses G1 method to determine the subjective weight, uses maximization deviation method to determine the objective weight, and according to principle of minimum deviation of the results of objective and subjective evaluation to determine portfolio weight is so as to establish science and technology comprehensive evaluation model based on portfolio weights. Third, it does empirical research of science and technology evaluation on typical provincial-level administrative regions.The contribution and special of this paper: First, it uses R cluster analysis to classify indicators in the same guideline, uses factor analysis to screen the indicator with the highest factor loading in each cluster, which both ensures the indicator selected has the greatest influence in the cluster it locates, and avoids information duplication in the same cluster. Second, it constructs a standard to determine whether the index system is reasonable based on how much information of sea-elected indicators has been reflected by screened indicators, and this idea exists in factor analysis that variance indicated that the idea of information content will be filtered reflected in indicators of information content, which solves the problem that indicator system's determination lack of quantitative criteria. Third, it establishes the optimization model based on the principle of Minimum deviation which is between the objective and subjective evaluation result, and then gets the weight proportion as well as the importance of objective and subjective evaluation results to the combination weight.