Spatial non-equilibrium and distribution dynamic evolution of the development level of national health in China’s provinces

Gini coefficient and its decomposition method

Based on the national health physique monitoring data of 31 provinces (cities, districts) in China in 2005, 2010 and 2015, the Gini coefficient and its decomposition method and nonparametric density estimation method were used to study the population, males, females, urban and rural nationals in China.The data underlying this article are available from https://www.sport.gov.cn/. These data were tested by specialized testers organized by the General Administration of Sports of China, all with the consent of the students and their guardians, and the data are publicly available and free of charge.

As one of the classic methods for the study of differences in various fields, the Gini coefficient and its decomposition method can overcome the problems that the Theil index and the traditional Gini coefficient cannot effectively solve the problems of overlapping and overlapping sample data and cannot further analyze the composition of regional differences and their sources (Dagum, 2008). 1997). In this paper, the Gini coefficient and its decomposition method are used to analyze the spatial disequilibrium degree, composition and origin of the development level of Chinese national health. Its specific definition is shown in formula (1): Among them, represents the overall level of the development of national health and physique in China, and the value is expressed by the average value of the development level of national health and physique in all provinces (cities and autonomous regions) in the country; y_ji (yhr)represents the development level of national health and physique in any province in the j(h)region, nrepresents the total number of provinces, k represents the number of divided regions, and n_j(n_h)represents the number of provinces in the j(h) region. In order to further analyze the composition and source of the spatial disequilibrium in the development level of national health and physique in the province, it is first necessary to rank the average development level of national health and physique in the region, as shown in formula (2): According to the decomposition method of Dagum (1997), G can be further decomposed into the sum of the contribution of intra-regional gaps (G_w), the contribution of inter-regional gaps (G_nb) and the contribution of hypervariable density (G_t), namely: The definitions of are shown in equations (4)-(6) respectively: Among them represents the Gini coefficient of the development level of national health in the j region, G_jh represents the Gini coefficient of the development level of national health and physique between h regions, D_jh represents the degree of influence between regions relative to the contribution rate of the national health and fitness development j(h)level,,the definitions are shown in equations (7)-(9)ο In the formula 9, d _jh is the difference in the contribution rate of the development level of national health and fitness between regions, p _jh is the hypervariable first-order moment, which is defined as formulas (10) and (11) respectively. Among them, F_j(F_h)represents cumulative density distribution function of area j hο

The Estimation Models of Nonparametric Density Kernel

In this paper, kernel density estimation is used to further analyze the dynamic evolution characteristics of provincial national health physique development level. Density estimation is a non-parametric statistical method for fitting the density function of sample distribution in statistical problems feature.

For the dataset{x₁, x₂ … x_n},the basic form of the density estimation Kernal function is Where is the h bandwidth, N is the number of observations, x is the average value of the observations, k(·)representing the kernel density, which is a weighting function or a smooth transition function, usually satisfying: The Kernel function mainly includes four types: Gaussian kernel, exponential kernel, triangular kernel and quartic kernel. In this paper, Gaussian kernel is selected to estimate the dynamic evolution law of the development level of national health in China’s provinces. Its function form is: After the function is estimated, the dynamic evolution trend of the variable under investigation can be judged by examining the distribution position, shape and ductility of the variable kernel density estimation curve.

Data sources and regional definition

This paper uses the comprehensive index of national health and physique in China’s provinces (cities and districts) to measure the development level of national health and physique in each province. The data mainly includes three aspects of body shape, physical function and physical quality, and involves five categories of national health and physical fitness index of 31 provinces (cities, districts), towns, villages, males, and females, which are very representative. Since the survey did not cover regions such as Hong Kong, Macau, and Taiwan, these three regions were not included in the analysis. In order to ensure the comparability of the data, all the data are processed uniformly with the first national health physique monitoring data in 2000 as the base.

Due to the large differences in economy, culture, infrastructure, etc. between regions, this paper further divides the 31 provinces into eastern (including 11 provinces) and central (including 8 provinces) according to the regional division standards of the National Bureau of Statistics), the west (including 12 provinces) three major regions.

Visual description of spatial disequilibrium in the development level of national health and fitness in China’s provinces

With the help of ArcGIS software, the spatial distribution map of the overall development level of Chinese national health in 2005 and 2015 was drawn, as shown in Figure 1 and Figure 2. From this, it can be seen that there is a significant spatial disequilibrium in the development level of national health and fitness in China’s provinces. Compared with 2005, in 2015, the number of provinces with a comprehensive index of national health and physique development level above 100 decreased significantly, and the number of provinces between 95 and 100 increased significantly. Although Shanghai was the province with the highest level of national health and physical development in 2005 and 2015, its comprehensive index of national health and physical fitness dropped by 1.67 compared with 2005. Jiangsu ranked second in national health and physical fitness level in 2005 (the comprehensive index was 105.7), in 2015, the national health and physique level dropped significantly (the comprehensive index was 102.92); and Zhejiang ranked second in the country with a comprehensive index of 104.99. In 2005 and 2015, the three provinces of Tibet, Qinghai and Guizhou ranked the bottom three in the country in terms of national health and physical fitness.

• Download figure
• Open in new tab

Figure 1

Heat map of regional distribution of the overall level of Chinese national health in 2005

• Download figure
• Open in new tab

Figure 2

Heat map of regional distribution of the overall level of Chinese national health in 2015

The composition and source analysis of the spatial disequilibrium in the development level of Chinese national health

In order to further analyze the regional disparities in the spatial distribution of the development level of China’s national health and fitness, its composition, and sources, we used the Gini coefficient and its decomposition method to calculate the overall, male, female, urban, and rural national health in 2005, 2010 and 2015, respectively. The Gini coefficient of the physical development level is further decomposed according to the three major regions of the east, the middle, and the west. The results are shown in Table 1-3.

Regional disparities in the spatial distribution of the overall level of national health and fitness and their sources

Table 1 shows the Gini coefficient and its decomposition results of regional disparities in the overall level of Chinese national health. In 2005, 2010, and 2015, the overall Gini coefficient of the development level of China’s national health and physique showed a trend of first decline and then rising, the lowest in 2010 (0.0167), and 0.07 and 0.1 percentage points higher than that in 2005 and 2010 respectively. It means that during the inspection period, the overall regional gap in the development level of Chinese nationals’ health and physique narrowed first and then expanded. In terms of sub-regions, the results of the Gini coefficient within the region show that the Gini coefficient of the overall level of national health and physique development in the eastern region shows a trend of first decreasing and then increasing, which is consistent with the changing trend of the overall national level; the Gini coefficient of the central region has increased from 2005 The Gini coefficient in the western region gradually decreased from 0.0205 in 2005 to 0.0118 in 2015, showing a downward trend year by year. It shows that there is obvious heterogeneity in the development trend of regional disparities in the level of national health and physique in the three major regions of China during the investigation period. The regional disparity in the development level of national health and physique in the eastern region first narrowed and then expanded, the regional gap in the development of national health and physique in the central region steadily expanded, and the regional gap in the development of national health and physique in the western region decreased year by year. From the perspective of regional disparities, the gap in the development level of national health and physique between the east and the west shows an evolution trend of first narrowing and then expanding, and its Gini coefficient is significantly higher than that of the east-central and central-west, which means that the overall level of national health and physique The east of the regional gap is the largest in the west [12]. From the comparison of East-Central and Central-Western, the regional gap between East-Central in 2005 (0.0115) was significantly lower than that of Central-Western (0.0184), while the regional gap between East-Central in 2010 and 2015 was higher than the Midwest. It shows that the regional disparity growth rate of the development level of national health and physique between the eastern and central regions is higher than that of the central and western regions. From the perspective of the source of the overall regional disparity and its contribution rate, the contribution rate of the inter-regional disparity is stable at more than 54%, and the contribution rate in 2005 and 2015 both exceeded 60%, indicating that the overall regional disparity in the development level of Chinese national health and fitness Mainly due to the gap between regions. The second is the intra-regional gap, whose contribution rate is stable between 24.78% and 27.15%, and generally shows a downward trend over time. The contribution rate of hypervariable density was the lowest, reaching the highest in 2010, at 19.75%, and in 2015, at the lowest, at 10.53%, showing a change characteristic of first increase and then decrease.

Regional disparities in the spatial distribution of male and female national health and fitness development levels and their sources

Table 2 shows the Gini coefficient and its decomposition results of the regional disparity in the development level of Chinese male and female national health. From the perspective of males, the regional Gini coefficient of the development level of Chinese male national health has increased from 0.0153 in 2005 to 0.0159 in 2015, showing an increasing trend year by year. From the perspective of the intra-regional Gini coefficient, the Gini coefficient of the development level of national health and physique in the eastern and central regions shows an increasing trend year by year, which is consistent with the changing trend of the overall level of men in the country; 0.0182 in 2005 gradually decreased to 0.0105 in 2015, showing a downward trend year by year.

From the comparison of the Gini coefficients of male national health and physical development levels in the three regions, in 2005 and 2010, the Gini coefficient in the western region was the largest, followed by the eastern region and the smallest in the central region; in 2015, the Gini coefficient in the eastern region was the largest, followed by the western region and the smallest in the central region. Judging from the Gini coefficient between regions, the gap in the development level of male national health and physique between the eastern and western regions showed an evolution trend of first decreasing and then increasing, and the gap was always higher than that of the eastern-central and central-western regions during the investigation period [13]. The Gini coefficient between the eastern and central regions showed a trend of first increasing and then decreasing, reaching a maximum value in 2010; the Gini coefficient between the central and western regions generally showed a downward trend. From the perspective of the source of the overall regional disparity and its contribution rate, the contribution rate of the regional disparity has been stable at more than 51%, indicating that the overall regional disparity in the development level of Chinese male national health is still dominated by the regional disparity. The second is the intra-regional gap, whose contribution rate is stable between 24.81% and 27.74%, and generally shows a downward trend over time. The contribution rate of hypervariable density is the lowest. During the investigation period, the highest value appeared in 2010, which was 21.67%, and the lowest value appeared in 2015, which was 13.87%.

From the perspective of women, the overall Gini coefficient of the regional gap and the Gini coefficient of the eastern region generally showed a trend of first decreasing and then increasing. From the perspective of the inter-regional Gini coefficient, the Gini coefficient between the eastern and western regions is still the largest. During the investigation period, its variation trend is similar to that of the central and western regions, showing a trend of first decreasing and then increasing. The Gini coefficient shows an increasing trend year by year. From the perspective of the source of the overall regional disparity and its contribution rate, the contribution rate of the inter-regional disparity is still the largest, accounting for more than 55% over the years, followed by the intra-regional disparity, but the contribution rate of the intra-regional disparity shows a downward trend year by year. The contribution rate of the over-editing density is the lowest, and the overall evolution trend during the investigation period showed an evolution trend of first increasing and then decreasing [14]. Compared with men, the overall regional disparity in the development level of female national health and physical fitness is significantly larger than that of men.

Regional disparities in the spatial distribution of urban and rural national health and fitness development levels and their sources

Table 3 shows the Gini coefficient and its decomposition results of the regional disparity in the development level of national health in urban and rural areas in China. From the perspective of cities and towns, the overall Gini coefficient of the regions with the development level of urban national health was basically stable at around 0.0185 in 2005 and 2010 and rose sharply to 0.0238 in 2015. The Gini coefficients within the eastern and central regions increased steadily year by year, from 0.0152 and 0.0075 in 2005 to 0.0374 and 0.0109 in 2015, respectively. The western region showed a gradual decline, from 0.0210 in 2005 to 0.0109 in 2015. From the perspective of the Gini coefficient between regions, the Gini coefficient between the eastern and western regions is still the largest, and generally shows a trend of first decreasing and then increasing. The Gini coefficient between the peaks reached 0.0193 in 2005, and was basically stable at around 0.0145 in 2005 and 2010. From the perspective of the source of the overall urban regional gap and its contribution rate, the contribution of the regional gap in 2005 and 2010 dominated, accounting for 58.26% and 44.86%, respectively. In 2015, the contribution of hypervariable density dominated, reaching 46.32%. On the whole, the contribution of the inter-regional gap showed a significant downward trend, while the contribution of the intra-regional and hypervariable density showed a steady increase, and the growth rate of the contribution of the super-density density was significantly greater than that of the intra-regional gap. From the perspective of the development level of rural national health and physique, the overall Gini coefficient of regional disparity and the internal Gini coefficient of the eastern region showed a trend of first decline and then increase, the central region showed a slight upward trend, and the western region showed a significant decline year by year. And by 2015, the internal disparity in the eastern region surpassed that in the western region [15]. It means that the evolution trend of the regional disparity in the development level of national health and physique in the eastern, central and western regions is first narrowing and then widening, slightly expanding year by year, and greatly expanding year by year. From the perspective of regional gaps, the regional gap between the east-west region is still the largest, and with the passage of time, the regional gaps between the east-west and east-central regions first increased and then narrowed, and the middle-Regional differences between the western regions are just the opposite. By 2015, the regional gap between the eastern and central regions surpassed that of the central and western regions, ranking second.

The dynamic evolution of the spatial distribution of the development level of Chinese national health

From the perspective of relative disparity, the above analysis analyzes the degree of spatial disequilibrium in the development level of Chinese national health and fitness, its composition, and sources. In the following, we further analyze the dynamic evolution trend of the spatial distribution of Chinese national health physique development level from the perspective of absolute disparity and spatial distribution form using a Kernel-density estimation method.

Kernel estimation of the dynamic evolution of the overall national health and fitness level

Figure 3 shows the distribution of the nuclear density curve of the spatial distribution of the overall national health and fitness development level in China from 2005 to 2015. From this, it can be seen that the distribution curve of nuclear density in 2005 presents a two-tailed multi-peak shape, indicating that the development level of national health and physique in some provinces is concentrated to a higher level, and the development level of national health and physique in some provinces is concentrated to a lower level, showing a “high”. The higher the higher, the lower the lower”. By 2010 and 2015, the distribution curve changed significantly: the height of the peak decreased significantly, the width of the peak increased, and the right-wing further extended slightly, showing the characteristics of evolution from a “spiky, multi-peak” shape to a “broad peak, single-peak” shape. It shows that while the overall development level of Chinese nationals’ health and physique has improved slightly during the inspection period, regional disparities have also further widened.

• Download figure
• Open in new tab

Figure 3.

Kernel density estimation diagram for the dynamic evolution of the overall national health physique level

Figures 4 and 5 are the distributions of the kernel density estimation curves of the development level of Chinese men’s and women’s national health and physique from 2005 to 2015, respectively. It can be seen from this that the dynamic evolution trend of the spatial distribution of male and female national health physique development levels has a strong similarity. From 2005 to 2010, the nuclear density estimation curves of the national health physique development level of both males and females showed that the peak value of the peak continued to decrease, the right tail slightly extended, the kurtosis gradually widened, and the evolution from a multi-peak shape to a single-peak shape was characterized. However, the nuclear density curve did not change much from 2010 to 2015 and remained basically stable. It shows that the regional disparity in the development level of male and female national health and physique has further increased on the whole. Among them, because the provinces with low levels of national health and physique of men and women show a certain “catch-up effect”, the polarization of low-level national health physiques of different genders gradually disappears, while the provinces with higher development levels of national health physiques The physical level has also been further improved. From the point of view of the peak position, the distribution curve of males remains basically unchanged, while the distribution curve of females shifts slightly to the left, indicating that the average level of Chinese male national health and physique has remained basically stable as a whole, while the average female national health physique has shown a slight increase. decline. From the comparison of the peaks, the peak value of the nuclear density curve of males far exceeds the level of 0.1, while the peak value of the nuclear density curve of females basically remains at the level of about 0.1, indicating that the absolute regional gap in the development level of male national health is generally smaller than women.

• Download figure
• Open in new tab

Figure 4.

Kernel density estimation diagram for the dynamic evolution of female national health physique level

• Download figure
• Open in new tab

Figure 5.

Kernel density estimation diagram for the dynamic evolution of male national health physique level

Figures 6 and 7 are the distributions of the kernel density estimation curves of the development levels in urban and rural areas of China from 2005-2015, respectively. It can be seen from the development level of national health in urban and rural areas that there is obvious heterogeneity in the dynamic evolution trend of the spatial distribution of the development level of national health and fitness in urban and rural areas in China. As far as towns are concerned, the distribution of nuclear density curves in 2010 and 2015 did not change much, but there was a big difference compared with 2005. With the passage of time, the kurtosis of the curve decreases and shifts slightly to the left, the right tail changes from an obvious single peak to a less obvious double peak, the left tail wave peak is further obvious and shifted to the left, and the left tail tip as a whole is greatly shifted to the left. Extend left. It shows that the regional disparity in the development level of national health and physique in China’s cities and towns has generally further increased, and some provinces with relatively low national health and physique levels are concentrated at lower levels, showing the characteristics of low-level “club convergence”; while high-level physique The trend of provincial concentration to a higher level has gradually eased. From the rural point of view, compared with 2005, the peaks of the curves in 2010 and 2015 decreased, the width of the peaks increased, and the tip of the right tail slightly extended outward, and the curves changed from multi-peak to single-peak. It shows that with the passage of time, the polarization of the development level of rural national health in China gradually disappears, and there is a “catch-up effect” in the physical development level of low-level provinces. The development trend of rural national health in different provinces tends to be Consistent.

• Download figure
• Open in new tab

Figure 6.

Kernel density estimation for the dynamic evolution of urban national health physique level

• Download figure
• Open in new tab

Figure 7.

Kernel density estimation diagram diagram for the dynamic evolution of rural national health physique level