Spreading knowledge and technology: Research efficiency at universities based on the three-stage MCDM-NRSDEA method with bootstrapping. (February 2022)

• Record Type: Journal Article
• Authors: Zhang, Chonghui; Jiang, Nanyue; Su, Tiantian; Chen, Ji; Streimikiene, Dalia; Balezentis, Tomas
• Is Part Of: Technology in society. Volume 68(2022)
• Publication Date: 2022-02
• Subjects: Research efficiency -- Three-stage data envelopment analysis (DEA) -- Multi-criteria decision-making -- Non-oriented super-efficiency DEA -- Bootstrap; Technology -- Social aspects -- Periodicals; 303.483
• DOI: 10.1016/j.techsoc.2022.101915
• Languages: English
• ISSNs: 0160-791X

Abstract

Abstract: Consequent to increasing higher education attainment and the expansion of higher education institutions, many countries have embarked on assessing discipline efficiency to track performance, promote competition, and ensure reasonable resource allocation. Therefore, measuring scientific-research efficiency is a crucial part of evaluating a discipline's development. This paper proposes the three-stage multi-criteria decision-making (MCDM) non-radial super-efficiency data envelopment analysis (NRSDEA) method with bootstrapping to study a discipline's scientific research efficiency from the university-level perspective. To ensure robust analysis, the proposed model incorporates the contextual variables describing the external environment and the random error. The data envelopment analysis (DEA) model is a non-oriented one. The three-stage DEA approach is applied, including contextual variables such as economic growth, innovation, infrastructure, and the natural environment. In addition, the bootstrap method is applied to correct for measurement errors. Finally, the research efficiency measurement of the statistics discipline at Chinese universities is taken as an example to verify the method's validity. Highlights: A three-stage multi-criteria decision making based NRSDEA method is proposed. Research Efficiency of the Statistics Discipline is assessed measured. The influence of environmental factors and random noise are eliminated by bootstrapping. The disciplinary perspective is considered to seek for the unique features of statistical research efficiency.