연구제안서 평가의 정확성에 관한 연구: 평가위원의 전문성과 평가자-피평가자 간 지인관계를 중심으로

In general, implementers of national R&D projects are selected by a scientific peer review system that appoints experts in the field. This study analyzed the accuracy of the research proposal appraisals by verifying the impact of evaluators’ expertise and acquaintanceship on the selection and implementation results of R&D projects, respectively. For this, we designed indicators to measure the Technology Evaluator Fit and the joint project implementation. An analysis of 19,447 projects submitted to the Korea Evaluation Institute of Industrial Technology over 10 years showed that the appraisal committee's expertise increased the accuracy of appraisals by reducing an underestimation bias. Meanwhile, it was confirmed that the acquaintanceship between appraisers and an appraisee, such as the joint implementation of past projects, increased the likelihood of selection regardless of the true quality of the project. The results of this study empirically support the assertion that the benefits of expertise outweigh the costs of biases except in the case of acquaintanceship, even though research proposal appraisals by a peer review method embody cognitive biases such as an in-group bias.