Exploring regional differences in mortality levels and their evolutionary trends is an important way to understand China's mortality and health transition. By using provincial mortality data from the third to the seventh censuses as revised by the Log-Quad Model and provincial statistics from various sources, this study analyzes variations in mortality levels and their driving mechanisms across age groups in China by space-time framework. It is found that firstly, the life expectancy gap between different regions in China has continued to show a pattern of being higher in the east and lower in the west, while the gap in infant mortality rates has narrowed and the gap in elderly mortality rates has increased. Secondly, with the shift in the epidemiological pattern of mortality, the driving mechanism behind the decline in mortality rates in China has shifted from medical facility investment to socio-economic development. The regional differences in mortality rates are largely driven by the stage-specific differences in the driving mechanisms of different regions. Given that socio-economic development has increasingly become the main driving force behind the convergence of regional mortality rates in China, efforts should be made to improve the socio-economic level of different regions, while consolidating existing medical investment.

Based on the 2015 Chinese 1% population survey, this study examines the spatial patterns of Chinese stem family and its influencing factors. It is found that there is significant heterogeneity among the distribution of stem families at the prefecture-level city with generally higher proportion of stem family in the south China and lower proportion in the north China. We further explore the spatial variation of factors affecting stem families using Geographic Weighted Regression (GWR) model. The results demonstrate that socioeconomic, demographic, and culture factors all play important roles in determining the distribution of stem families. The former two factors can exhibit opposite impact conditioned on the local culture and the development model of urbanization. Meanwhile, the impact of housing price is also divergent depending on the sensitivity of local housing market to the price change. These results suggest that the interplay of socioeconomic level, development mode, and cultural tradition ultimately shapes the pattern of Chinese stem family.

This study explores how to use the online public opinion big data to infer the transition of fertility intention propensity of Chinese web-users. By comparing the results from the simple sentiment analysis, the LDA un-supervised classification, and the fasttext supervised classification, it is found that the supervised classification based on the theory of planned behavior yields more accurate and meaningful results. A change from the positive to negative dominant mood of fertility intention has occurred since 2012, during which there is a shift of topics. The comments regarding perceived controls have replaced those related to behavioral beliefs as the most popular ones for fertility. Housing price, children's education, and work are the most mentioned key words in the category of perceived controls. The strong correlation between fertility intention propensity and fertility level among Chinese provinces validates our fertility intention analyses. We also discuss the prospect of applying public opinion big data in demographic research.

Utilizing the Chinese Health and Retirement Longitudinal Survey ,this study constructs the measure of biological age of Chinese elderly and tests its validity. In comparison to other health measures used in survey,the biological age shows several prominent advantages in the following aspects.First,the biological age,as a comprehensive measure of people’s physiological condition,can well predict individual’s mortality risk and health status in various dimensions. Second,it can reflect the diverging health status among the middle age and younger old-age people when many health problems have not been discovered. Third,compared with the subjective health measures,the biological age can eliminate the bias of reference point by different groups. Finally,the influence of chronological age on health can be eliminated from the biological age model. Based on the hierarchical linear regression model,it is found that education level is the most reliable factor that affects individual’s biological age and cohort can significantly differentiate old people’s health. In particular,individuals born during the great famine period may be confronted with a greater physiological risk.

Using data from CGSS 2003 ～2013 this paper explores the dynamic relationship between parenthood which is characterized by children’s number and sex structure and parents’subjective well-being．Through the hierarchical APC model，we derive the changing trends of such relationship along the time dimension of age，period and cohort．It is found that the effect of children’s number on parents’subjective wellbeing changes with parents’age．Having more children would bring negative effect to their young or middle aged parents but significantly promote the parents’happiness at their old age．Such age-differential effect is more salient for mothers．The happiness effect of children’s number also differs among parents’cohorts．Parents who were born in the 1940s enjoy the maximum happiness return from their children，after the cohort of which the positive effect rapidly diminishes due to the birth control regulation and demographic transition．Meanwhile，the happiness effects of having son differ between urban and rural parents．While having son slightly promote the subjective well-being of ru- ral parents，it significantly decreases the well-being of urban parents after their middle ages．Finally having daughters would marginally increases their parents’wellbeing，and the effect neither changes with parents’age or cohort nor differs between rural and urban mothers and fathers．

In this work, we utilize the vitality model to fit the agespecific mortality data of Chinese adult (35+), and decompose the adult life expectancy into the intrinsic life expectancy and the lost life expectancy due to extrinsic mortality. By applying the method to the Sixth Population Census data, we find that the gender difference in the Chinese adult life expectancy primarily stemmed from their difference in the lost life expectancy due to extrinsic mortality, while the difference between the urban and rural residents mainly reflected in their intrinsic life expectancies which were determined by the chronic aging process. The comparison with Japan further reveals that the disparity in the intrinsic life expectancy contributed mostly to the gap of these two countries in the adult life expectancy. Furthermore, the longitudinal comparison between the Fifth and the Sixth Census data suggests that the reduction of extrinsic mortality was the major force that enhanced the male life expectancy in the past ten years, while the decline of both intrinsic and extrinsic mortality characterized the change of the female life expectancy in the same period. According to the experience of countries with advanced life expectancy, the growth of the life expectancy in China will be dominated by the intrinsic increase.

Understanding the changing patterns of the health indicators of older population and the influencing factors is important for correctly estimating the healthcare burden and promoting the life quality for older population. Previous studies based on cross-sectional or single birth cohort data cannot distinguish or control the cohort effects，which may lead to biased estimation of the trend. Using the multi-cohort data of Chinese Longitudinal Healthy Longevity Survey ( CLHLS ( 1998—2011) ) ，this study investigates the growth curve trajectories of older people's self-rated health，Activities of Daily Living ( ADL) ，psychological wellbeing，and cognitive score; and explores their rural-urban disparities with full consideration of the cohort effects. The results suggest that the changing patterns of these health indicators and their population disparities are closely related to the cohort variations: ( 1) without controlling
the cohort effects，the estimation of age patterns would be biased; ( 2) there are cohort variations among the health trajectories; ( 3) within the same cohort，urban and rural elderly show significant differences in all of the health indicators; and ( 4) such differences in ADL and psychological wellbeing
change with cohort.