Abstract: Both public foundations and private foundations are important forces in scientific research funding, and both have made great contributions to the development of scientific research in different historical stages. The analysis of the differences between public foundations and private foundations is the basis of an effective national scientific research funding layout design. The topic has drawn the attention of both overseas and domestic scholars: Overseas scholars tend to adopt quantitative research methods, using funding data sample to analyze the differences between the two types of foundations; whilst domestic researchers focus more on quantitative methods to investigate in the traits of private foundations funding, yet most of which ignore the continuity and completeness of data sample, the diversity of views of research, in particular the combination of funding input data and research output data and synthesis of traditional analytical methods and newly-emerged technology. Therefore, this article selects multi-dimensional information contained in the funding projects and output papers funded by public and private foundations in the UK biomedical field from 2009 to 2018 as the data sample to construct a multi-dimensional funding difference analysis model. Besides the traditional statistical analysis and bibliometric methods, the paper also introduces a content analysis method based on machine learning algorithms.In a more detailed manner, the paper finds out that the differences between the two types of foundations lay in 3 aspects on funding input level: a. Public foundations enjoy a larger gross funding amount than private foundations; b. Public foundations tend to support research projects with a more conventional funding cycle, usually between 2 to 4 years, while private foundations tend to support projects either less than 1.5 years or more than 4.5 years; c. Public foundations tend to support research topics within a regular funding scale, usually between 300,000 to 1 million pounds, while private foundations tend to support research with either smaller or larger funding amount, usually less than 250,000 or more than 1 million pounds. In terms of research output performance, the study discovers the differences between the two types of foundations in3 ways: a. Private foundations show a bigger size of research teams and varieties than public ones; b. Private foundations present a better degree of research topic novelty than public foundations in terms of amount, proportions and time sequence; c. Private foundations enjoy a better research output performance than public ones in both academic output and impacts on technology innovation. Based on a long-term, large sample, and multi-dimensional analysis, the following conclusions have been drawn: a. Public foundations and private foundations have various funding priorities and complement each other. In recent year, both foundations are keeping searching for and adjusting their positions, in the sense that public foundations tend to be more 'conventional' in funding cycle project organization and project contents, yet private foundations appear to be more 'adventurous'. b. Private foundations have formed beneficial practices in promoting cooperation, improving academic performance, and serving technological innovation and development, which public foundations are worth learning from. c. The funding from public foundations is accused of "keeping old wine in new bottles", in the sense that public foundations tend to seek novelty through offering new names or creating fancier words to the same research topics, yet the output performance remain less surprising; While the funding from private foundations enjoy the effects of "old trees growing new branches", as meaning that in private foundations, though keeping the original funding framework, new research topics and directions are developed in an interesting manner. The above conclusions provide a basis for a more comprehensive and in-depth understanding of the funding differences between the two types of foundations, and also lay the groundwork for the rational and effective deployment and design of government scientific research funding. Finally, the paper offers three policy suggestions: a. The government shall learn from the private foundations and improve the current state funding system; b. The government shall set up a package of incentive mechanisms and encourage more funding from private foundations; c. Besides private foundations, entrepreneurial, provincial and municipal funding shall also be taken into the state research funding systems.
Keyword：public foundations; private foundations; funding science; difference;
文章作者：阿儒涵 王悦 李晓轩