Abstract:

Optimizing the spatial distribution of talent resources plays an important role in achieving coordinated regional development. Based on the county data of the two national censuses in 2010 and 2020, the spatio-temporal pattern of China's highly skilled talents is sketched during the 10-year period, and a Beta regression model is used to identify the influencing factors of the spatial distribution of highly skilled talents. The results show that: ①Hu Huanyong line is a significant demarcation line for the distribution of high-skilled talents, and the proportion of high-skilled talents in the northwest side of Hu Huanyong line is generally higher than that in the southeast side, and it is relatively stable in ten years. ②The proportion of highly skilled talents in the southeast side of the Hu Huanyong line shows a "core-edge" spatial structure decreasing from the city district to the surrounding neighboring counties, while the northwest side of the line generally shows a homogeneous character, and the spatial differentiation is not significant; The proportion of highly skilled talents shows a hierarchical differentiation from municipalities, provincial capitals, and municipal districts of planned cities to prefecture-level municipal districts, county-level cities, and counties, decreasing step by step; The spatial pattern of the proportion of highly skilled talents in China shows a generally stable and locally changing trend, and the changing counties are mainly concentrated in the Northeast, the Yangtze River Delta, Tibet and other regions; ③The administrative level of counties, GDP per capita, urbanization rate, non-farming employment, proportion of tertiary industry, average altitude, temperature difference, and NDVI together affect the distribution pattern of the proportion of highly skilled talents.

Key words: highly skilled talents, spatio-temporal pattern, exploratory spatio-temporal data analysis, Beta regression model, China

摘要：

人才是创新驱动发展的核心要素，优化人才资源的空间布局，对于实现区域协调发展具有重要作用。基于2010年和2020年两次全国人口普查分县数据，采用探索性时空数据分析方法刻画10年间中国高技能人才的时空格局，并采用Beta回归模型识别高技能人才空间分布的影响因素。结果表明：①胡焕庸线是高技能人才分布的分界线，胡焕庸线西北半壁高技能人才占比普遍高于东南半壁，且在十年内相对稳定；胡焕庸线东南半壁高技能人才占比呈现出由市辖区到周边县域递减的“核心-边缘”结构，而西北半壁总体呈现均质化分布特征，空间差异不显著；②高技能人才占比呈现由直辖市、省会城市、计划单列市市辖区，向地级市市辖区，县级市，县（旗、行委）逐级递减的等级分异特征；③中国高技能人才占比空间格局呈现总体稳定、局部变动的态势，变动县域主要集中在东北地区、长三角、西藏等区域；④县域的行政等级、人均GDP、城镇化率、非农就业机会、第三产业占比、平均海拔高度、温差、NDVI共同影响高技能人才占比的分布格局。

关键词: 高技能人才, 时空格局, 探索性时空数据分析, Beta回归模型, 中国