Scientific effects of large research infrastructures in China

•This paper firstly evaluates the scientific effects of RIs in China by multi-case analyses.•This paper proposes a comprehensive analytical framework for the scientific effects of RIs.•This paper presents some evidence for the active role of RIs in the scientific performance.

Large research infrastructures (RIs) are expected to play an important role in the development of scientific activities in China and the construction of China's national scientific systems. However, few studies have been devoted to the systematic evaluation of the scientific effects of China's RIs. This paper attempts to fill this gap by designing a comprehensive analytical framework composed of the input-side, output-side, process-side and environment-side effects of RIs on scientific activities. The analysis is implemented based on a novel sample composed of nine of China's typical RIs. More specifically, this paper classified these nine Chinese RIs into the following three types according to their functions: dedicated research infrastructure, public experimental platform and public infrastructure. Furthermore, this paper analyzes the features of the scientific effects of these RIs in terms of the following four typical scientific effects: science and technology advancement effect, capability cultivation effect, networking effect and clustering effect. In addition to the finding that RIs have promoted scientific advancements in many disciplines in China, the study found that RIs are important to the acquisition of new knowledge, and also contribute to the propagation of competitive scientific organizations and scientific talent. Networking and clustering impacts are also important scientific effects of RIs, as they increase the effectiveness of scientific activities in China. This paper not only contributes to developing an analytical framework for evaluating the functions and effects of large RIs but also presents evidence regarding the development of large RIs in emerging countries.