Abstract:     Transformation of scientific and technological achievements is a crucial undertaking for universities to contribute to China′s economic and social high-quality development. How to enhance the transformation efficiency stands as an urgent need to study. Within the research framework of proximity and innovation, this paper made an estimation of the efficiency of knowledge production and industrial transformation, which are two stages of university′s transformation of scientific and technological achievements. To this end, the bootstrap DEA model was employed, using the data of universities and cities within urban agglomerations of the Beijing-Tianjin-Hebei, the Yangtze River Delta, and the Pearl River Delta from 2006 to 2017. Subsequently, the sample selection model was used to analyze the influence of geographical proximity of both the intra- and inter-urban agglomerations on universities′ efficiency, with regards to two stages of scientific and technological achievements transformation.The results of this study are as follows. Both knowledge production and industrial transformation in universities are inefficient within the three urban agglomerations. The universities perform better at the stage of knowledge production but face more challenges when they come to the stage of industrial transformation. After considering the evolution of the Internet, the spatial connection of intra-urban agglomeration has a significant positive effect on the transformation efficiency of scientific and technological achievements at both stages, which means geographical proximity cannot be wholly replaced with Internet connection. In addition to other factors, the number of associate professors and professors, as well as the universities with the prestigious title of "first-class universities and disciplines of the world", improve knowledge production efficiency significantly. However, they are not conducive to industrial transformation. At the stage of knowledge production, the funds allocated by enterprises and institutions promote universities′ efficiency significantly, whereas government grants produce the opposite effect. Nonetheless, at the stage of industrial transformation, both types of funding have a positive impact on universities′ efficiency. The findings will bear significant practical implications for advancing the seamless integration of urban agglomerations, fostering innovative environments, directing diverse investment and focusing on industry demand at the knowledge production stage, as well as developing holistic policy and interdisciplinary collaboration during the stage of industrial transformation.

Key words: transformation efficiency of scientific and technological achievements, geographical proximity, Bootstrap-DEA, sample selection model