Global multidimensional poverty has arouse great attention from the public. It is of great theoretical and practical significance to explore the spatial and temporal changes of multidimensional children poverty and its influencing factors. Taking 25 provinces (municipalities and autonomous regions) of China as an example, this study uses the A-F method to measure children multidimensional poverty of China during 2010-2016, and examines its spatio-temporal pattern and influencing factors with spatial autocorrelation and the geodetector. The results demonstrated the following: (1) During 2010-2016, the children multidimensional poverty index in eastern, central and western China showed a downward trend, and poverty in the dimensions of living standards, care, education and health were effectively improved. (2) The children multidimensional poverty index and each dimensional poverty index had spatial differences, which was manifested as "high in the east, medium in the central region, and low in the west". (3) Urban and rural children multidimensional poverty in China has been improved, and the index spatial pattern has gradually changed to "east, middle, west; high, middle, and low", however, the multidimensional poverty index of rural children has been far higher than that of urban children. (4) Family environment and urbanization level are important factors affecting children multidimensional poverty. Ability of raising children, economic level and educational environment are secondary factors. The interaction between the factors has a far greater impact on children multidimensional poverty than a single factor. Educational environment and ability of raising children, urbanization level and medical resources, and family environment are the main interactive factors leading to children multidimensional poverty.

As a "package of projects", the targeted poverty alleviation strategy has accelerated economic, social and spatial reconstruction of rural areas. This paper aims to conduct a theoretical analysis of the connotation, characteristics, process, and mechanism of targeted poverty alleviation that drives poor rural reconstruction and provide empirical evidence of Shibadong Village, which is the first initiative of targeted poverty alleviation. It is found that targeted poverty alleviation is a process of multidimensional, comprehensive, and rapid reconstruction that is led by the government with multi-party participation in combination of "top-down" and "bottom-up" approaches. The process of targeted poverty alleviation changes the poor rural areas from 'low-level poverty equilibrium' to 'healthy development equilibrium' through a combination of administrative power, sending elites to the village level, improving market accessibility, and coordination of factor integration, motivation, planning, training and institutional guarantee mechanisms. Since the implementation of the targeted poverty alleviation, Shibadong Village has transformed from a dualistic economy to a pluralist economy, from a pure acquaintance society to a semi-acquaintance society with a mixture of acquaintance logic and contract spirit, and from a spatial form highly overlapped from the production and living spaces, yet separated from the ecological space to a diverse and orderly cluster of organic integration of production, living and ecological space. The practice of Shibadong Village provides a fresh sample for us to understand the process and mechanism of the restructuring of poverty-stricken villages, and it is also an essential embodiment of its "reproducible and extendable" experience.

Market spatial correlation is one of the key indicators that reflect the operation efficiency of corn markets. The implementation and reform of corn price support policy in China plays an important role in changing the spatial correlation of corn markets. Employing market price data at province level, this study first analyzed the change of corn market spatial correlations under the price support policy reform. Then, social network analysis was applied to describe the network characteristics of corn market spatial correlations, and quadratic assignment procedure (QAP) analysis was used to investigate key factors that affected the spatial correlation. The results indicated that, (1) overall, the spatial correlation of China's corn markets was relatively high, and there was an upward trend in recent years. The implementation of temporary corn storage policy exerted insignificant effect on this upward trend, but it led to a decrease in the spatial correlation between markets in policy-affected provinces and those in non-policy affected provinces. (2) The spatial correlation of corn markets presented a multi-threaded and complex network structure, which was closely related and stable as a whole. Provinces with large corn consumption in the central and eastern regions were at the center of the spatial correlation network and played as the central actor in the network. However, provinces in the western region and provinces implementing corn support policy had little effect on the spatial correlation network of corn markets, indicating their marginal and disadvantaged role in the network. (3) Geographical proximity, market distance, market power, information transmission effect and the implementation of temporary storage policy were the key factors that affected the spatial correlation of corn markets. Furthermore, the role of market power and information transmission effect in determining the spatial correlation has become increasingly important.