Peer-reviewed Specialist Anti-assessment Model

Expert anti-assessment means the evaluation of the ability and level of experts in science and technology work, according to the activities they took part in. Expert anti-assessment can provide the expert selection of decision support, make the science and technology evaluation activities more objective, fair and impartial, play the role of supervision and incentives to promote their correct attitude, and cause the attention of relevant departments of science and technology management. This thesis established an expert anti-assessment model and provided a new expert anti-assessment method.This thesis analyzed the science and technology evaluation data of various science and technology management departments in recent years, and brought the outlier detection theory into the study of expert anti-assessment. Grubbs's test was used to detect the outliers of science and technology data experts gave to the projects in science and technology activities. After excluding the outlier data, the averages were calculated from the remaining data, and regarded as the basis of expert anti-assessment. Apart from this, expert anti-assessment model based on distance and correlation coefficient was proposed. This model mapped the range of distance and correlation coefficient to three different discrete values respectively, according to their own characteristics. After calculating the "distance" and "correlation coefficient" between the data of the projects expert evaluated to and the average evaluation values, we obtained the two corresponding discrete values of the two statistics, and added the two discrete values as the ability and level of the experts in science and technology assessment activities.Experimental results show that the expert anti-assessment model based on distance and correlation coefficient proposed in this thesis achieved the goal of evaluating the ability and level of the expert in science and technology activities, and it has been recognized by several relevant science and technology management departments.