A highly educated workforce is an important driving factor for urban innovation and development. A better understanding of spatial patterns and location determinants of highly educated workers is valuable for designing policies to attract them. The existing literature has investigated the impacts of economic variables and urban amenities on the patterns of highly educated workers. However, the relationship between urban amenities and highly educated workers needs to be investigated further. Moreover, few studies have examined the impact of urban size on the agglomeration of highly educated workers. Thus, this study employed a spatial analysis method to examine the pattern of highly educated workers across 300 cities in China, based on 2010 census data. We used a negative binomial regression model to identify the determinants of highly educated workers' locations. The results are numerous. First, the size of highly educated workers varied greatly among cities of different ranks, and its rank-size distribution was consistent with power function characteristics. It was characterized by spatial convergence and zonal differentiation from East to West. Highly educated workers were distributed mostly across the eastern coastal area, the central-southern area of Liaoning peninsula, the Harbin-Changchun urban agglomeration, and in inland provincial capitals. Second, the share of highly educated workers in total employment in the resource-dependent cities of north China and eastern coastal urban agglomeration was significantly higher, while it was obviously low in central and southwest China. Comparing the rank-size distribution characteristics of highly educated workers along with their total employment, the cities with rich attractions for highly educated workers were mainly located in Inner Mongolia, Northeast China, the Beijing-Tianjin-Hebei metropolitan area, the Shandong Peninsula, the Yangtze River Delta, the Pearl River Delta, and inland provincial capitals. Third, the estimated results show that the spatial pattern of highly educated workers was affected by urban amenities, economic opportunities, and the agglomeration effect. At the national level, economic opportunities, city size, and housing prices dominated the spatial pattern of highly educated workers, while the cool summer climate, high quality education resources, and recreation and transportation resources also played important roles. Fourth, the distribution of highly educated labors was affected by both economic opportunities and amenities, but the role of amenities was quite different in different size-ranked cities. The main amenity factors for the disparities of highly educated workers in large cities were found to be those of temperature in January and July, high-quality early childhood education resources, recreation resources, and traffic conditions. For the medium-sized cities, the dominant amenities included high-quality primary and secondary education resources and cultural and artistic services. Regarding the disparities of highly educated workers among small cities, the main contributing amenity factors were high-quality preschool and middle-school education resources, high-quality medical resources, recreation resources, and air quality. This study supports previous opinion that the amenities contribute to attracting talent. It also reveals the role of the agglomeration effect in the spatial distribution of the highly educated labors, which enriches our understanding for the location choice of highly educated workers in economic geography. This study enlightens us in the sense that the policies to attract talent should not only take labor market into account but also work to improve the amenities that such people work in.