Spatial Demography最新文献

Model migration age schedules have proved valuable to demographers for a range of applications for over 40 years. The original Rogers-Castro curve has been extended over time to include a retirement curve, a post-retirement curve, and a student peak. With demographic analyses extending to higher age groups than in the past due to population ageing, it is important for the model schedule to faithfully reflect migration patterns at advanced ages. Recent data on internal migration in the nonagenarian and centenarian ages reveals several examples of rising then falling mobility with increasing age. This paper suggests an alternative specification of the post-retirement curve of the model schedule to reflect this pattern. The modified model migration schedule is successfully fitted to example internal migration age patterns from Australia, Canada and the Netherlands. The modified schedule should prove useful in preparing input data for population projections and analyses of migration age patterns extending to the highest ages.

Amid a persistent U.S. fertility decline since the Great Recession, fertility recuperation patterns by geographic regions were not homogeneous. This study hypothesizes that the geographic discrepancies in fertility patterns are attributable to different labor force compositions by the regions. We use data from the U.S. Census Bureau’s Annual Estimates of the Resident Population County Components of Population Change to estimate the discrepancy in fertility variations at the county-level. By comparing the slopes of births before and following the recession, we visualize the characteristics of fertility variations at the U.S. county-level. Also, a multiple linear regression model estimates that the counties with a greater share of labor force in wholesale trade, information & technology, finance & insurance, and professional & scientific industry show greater volatility in fertility trends throughout the Great Recession. On the contrary, the counties with higher proportions of the labor force in agriculture, retail trade, and education industry tend to less change over the years of the economic recession. However, fertility recuperation is limitedly identified amid the structural fertility decline after the Great Recession.

Changes in the spatial patterns of ethnic diversity and residential segregation are often highly localized, but inconsistencies in geographical data units across different time points limit their exploration. In this paper, we argue that, while they are often over-looked, population grids provide an effective means for the study of long-term fine-scale changes. Gridded data represent population structures: there are gaps where there are no people, and they are not (unlike standard zones) based on population distributions at any one time point. This paper uses an innovative resource, PopChange, which provides spatially fine-grained (1 km by 1 km) gridded data on country of birth (1971-2011) and ethnic group (1991-2011). These data enable insight into micro-level change across a long time period. Exploring forty years of change over five time points, measures of residential ethnic diversity and segregation are employed here to create a comprehensive 'atlas' of ethnic neighbourhood change across the whole of Britain. Four key messages are offered: (1) as Britain's ethnic diversity has grown, the spatial complexity of this diversity has also increased, with greater diversity in previously less diverse spaces; (2) ethnic residential segregation has steadily declined at this micro-scale; (3) as neighbourhoods have become more diverse, they have become more spatially integrated; (4) across the whole study period, the most dynamic period of change was between 2001 and 2011. While concentrating on Britain as a case study, the paper explores the potential offered by gridded data, and the methods proposed to analyse them, for future allied studies within and outside this study area.

Poverty has been studied across many social science disciplines, resulting in a large body of literature. Scholars of poverty research have long recognized that the poor are not uniformly distributed across space. Understanding the spatial aspect of poverty is important because it helps us understand place-based structural inequalities. There are many spatial regression models, but there is a learning curve to learn and apply them to poverty research. This manuscript aims to introduce the concepts of spatial regression modeling and walk the reader through the steps of conducting poverty research using R: standard exploratory data analysis, standard linear regression, neighborhood structure and spatial weight matrix, exploratory spatial data analysis, and spatial linear regression. We also discuss the spatial heterogeneity and spatial panel aspects of poverty. We provide code for data analysis in the R environment and readers can modify it for their own data analyses. We also present results in their raw format to help readers become familiar with the R environment.