In urban areas, traffic emissions are a major source of air pollution, which results in health effects. Traffic-related pollutant concentrations vary both horizontally and vertically with distances from major roads, resulting in different exposure levels for urban dwellers. The objectives of this study are (1) to assess the three-dimensional distribution of the population with high exposure to traffic emissions at the district level with an innovative approach, Three-Dimensional digital Geography methodology (3DIG), and (2) to explore the socio-demographic characteristics of the high exposure population in Taipei. The results show that on average 14% of people live within 5 m of municipal roads on the first and second floors in each township in the studied area. High Exposure Population Density (HEPD) represents the number of people within the high exposure areas per square kilometer. After excluding one extreme value, the average HEPD is 3,771 persons/km^2 with a standard deviation of 4,006 persons/km^2; the maximum is 38,519 persons/km^2. This clearly shows the spatial inequality of traffic exposure in Taipei. Additionally, a correlational analysis indicates that females aged 15 to 64, and those with higher income and higher education tend to live in the higher HEPD areas. Furthermore, geospatial analysis shows that the population with high traffic-related pollution exposure is highly clustered, resulting from the expansion of metropolitan Taipei. However, the socio-demographic spatial inequality does not coincident with spatial inequality of traffic exposure. It turns out that residents with higher education levels living in downtown areas have relatively high traffic pollution exposure in Taipei. This study demonstrates that 3DIG is a versatile methodology which can be used in any research focusing on three-dimensional population distribution.

Drawing from theories of son preference and family resources, this paper attempts to construct a sequential fertility model, and then uses it to estimate the probability of parents' fertility decisions for their first few children by the sibling structure and to elaborate the effects of various socio-demographic factors. The data used in the analysis come from the PSFD, in which respondents were born between 1934 and 1986. The main findings are as follows. Firstly, the sibling structure has a significant effect on fertility decisions. Secondly, the gender-balance hypothesis is rejected, whereas the son-preference hypothesis is supported. Parents with daughters in their first two or three births have a greater likelihood to have a next child. Thirdly, son preference and son investment have opposite effects on fertility decisions. The number of sons tends to reduce the likelihood of parents having another child. Fourthly, the hypothesis of social stratification is rejected. Parents of different status groups make similar fertility decisions. Fifthly, the data didn't support the cohort hypothesis, because the patterns of fertility decisions by different cohorts appear the same. Finally, the significant effect of sibling structure indicates that fertility decisions are made in a sequential decision-making process.

Long-term care services can be divided into informal care and formal care. Family care, the informal one, is not only more common, but also the preferred mode for elderly people in Taiwan. According to the literature, preference for family care may affect the demand for long-term care insurance. This study investigated the family care preference, as well as other factors that affect the demand for long-term care insurance in Taiwan. The results are as follows: 1. Preference for family care did not affect the demand for long-term care insurance, and the availability of family care such as children and spouse had no statistically significant effect on the demand for long-term care insurance. 2. Middle-aged people had higher demand for long-term care insurance. 3. Respondents who earned less than NT80,000 per month had higher demand for long-term care insurance, although they had limited ability to pay premiums. 4. Middle-income respondents had higher demand for long-term care insurance, as long-term care insurance acts as an asset protection instrument. 5. The perception of the long-term care issues and education also affected the demand for long-term care insurance.

The effect of spatial concentration and hierarchy of economic activity on national growth is investigated using cross-section OLS and dynamic panel GMM estimation. Agglomeration patterns are measured by urbanization shares and level of urban hierarchy, which is measured through rank-size coefficient. We find evidence supports the ”Williamson hypothesis”; however, the critical level is estimated at USD 7,100 lower than that of previous work. The impact of the level of urban hierarchy on country level growth is significant and positively related to national income. This result implies that not only the degree of absolute agglomeration effect but also the relative size of agglomerations among cities is worth considering.