Abstract: The science and technology service industry is key to promoting the organic integration of technology and the economy. Describing the spatial distribution pattern of the science and technology service industry within a city and analysing influencing factors of location selection have essential theoretical and practical significance to reflect spatial distribution patterns of the industry and achieve the optimal allocation of science and technology resources. From a considered and extensive literature review, associated research in China is primarily concentrated in eastern cities, such as Beijing, Shanghai and Hangzhou. There are few research laboratories in cities in Central China. For research content, most studies focus on the integral performances of the science and technology service industry. However, there is a lack of discussion on the differences in the spatial pattern among different types of science and technology service industries.
　　Moreover, influences of factors, such as cultural environment, incubation environment and business environment, on the location selection of the industry were ignored. Comparative studies on the influencing factors of different enterprise types have also rarely been addressed. Research methodologies often apply qualitative description analysis, resulting in limited studies using quantitative analysis. Based on data from industrial and commercial registered enterprises in Wuhan, the spatial pattern of the science and technology service industry and the location differences among different types of enterprises in the Wuhan Metropolitan Area were investigated through circle analysis and kernel density estimation. The negative binomial regression model was used to quantitatively analyse location selection factors and identify differences among different industries. The Wuhan metropolitan area's science and technology service industry concentrates in central and suburban areas. Two peak areas are identified in the 3-6km and 15-18km circles. The "multi-centre" spatial pattern has been formed. The core agglomeration areas are Optics Valley Chuangye Street, Guandong Science and Technology Park, and Harbour of Technology in Optics Valley. The secondary agglomeration areas are the Zhongnan Road business district, Jiedaokou business district, Optics Valley Software Park, and Wuhan University Science and Technology Park. The Wuhan Metropolitan Area's science and technology service industry is mainly distributed in Venture Street, Science and Technology Park, and Headquarters Base.
　　Commercial centres, university clusters and software parks are also attractive, and there are apparent differences in the spatial pattern among different industries. Research and experimental development companies present a spatial pattern of "one principal and two associates", tending to be located in neighbouring science and technology parks. Professional technical service companies present a spatial pattern of "two principals and multiple assistants", tending to be close to university clusters and commercial centres. Technology promotion and application service companies present a spatial pattern of "one master, multiple assistants" and distribute along with commercial centres. The influence of location conditions, business environments, technical factors, cultural environment, incubation environment, agglomeration factors, and policy factors primarily influences the study area's spatial patterns. Comparing various influencing factors resulted in identifying that technological factors, incubation environment and policy factors are primary influences on the location selection of the industry. The science and technology service industry focuses on resource elements, such as technology, talents, incubation and policies. The influencing factors have different effects on different types of industries. Compared with other companies, research and experimental development companies pay more attention to technical factors, cultural environment and policy factors, with no particular inclination to consider the business environment. For professional technical service companies, the effects of distance from subway stations and land prices are higher than that of other industries. Technology promotion and application service companies are more sensitive to changes in a business environment, incubation environment and agglomeration factors than other companies. These companies are relatively weakly affected by the cultural environment. 

Key words: science and technology service industry, spatial pattern, influencing factors, negative binomial regression model, Wuhan metropolitan area

摘要： 城市科技服务业区位选择的研究对于科技资源的优化配置具有重要意义。以武汉都市区为案例，基于工商登记企业数据，运用圈层分析与核密度估计方法，刻画城市科技服务业的空间格局，利用负二项回归模型检验其区位选择的影响因素。结果表明：城市科技服务业集中分布于中心区和近郊区，呈现“多中心”的空间格局，不同类型科技服务业的空间格局存在明显差异；区位条件、商业环境、技术因素、文化环境、孵化环境、集聚因素和政策因素是科技服务业空间格局的主要影响因素，影响因素对不同类型科技服务业的作用强度存在差异。

关键词: 科技服务业, 空间格局, 影响因素, 负二项回归模型, 武汉都市区