Spatial demography

The issue of changes in household size and composition in developing world is attracting scholarly attention only in recent years . Such changes have been quite widespread in economically developed countries. Demographers have in the past neglected the quantitative dimensions of the size composition and change in households and their causes and consequences. Changes in household size are clear reflections on changes in living arrangements as well as in household composition. The problem acquires nuances of meanings in a country like India which has shown indications towards a tendency to move into a low fertility regime in the years to come. Aggregate demographic statistics generally mask substantial changes in the size and number of households. As in most indicators of demographic, economic and social changes, household changes are neither uniform in their social and spatial content. The paper examines inter-state variation in this feature of change that is unfolding in India in recent years. The magnitude of household changes-both spatial and social-constitutes central problematic of the research.

Although staying - even more than moving - is a common feature of human existence, the nature of staying is heavily under researched. Numerous studies have given us extensive knowledge about migration and migrants, however much less is known about the non-migrants and their characteristics and decision processes. This study seeks to address this gap in the literature by focusing on the specific nature of staying. At the centre of attention are Swedish life-long stayers aged 60 +. By the use of the Linnaeus database, micro-level population data covering the full Swedish population and containing apart from rich socio-economic data and links between generations also detailed geographical information, the quantitative studies of the extent and patterns of staying can be addressed. By utilizing population register the life course and characteristics of the “stayers” are scrutinized, and questions concerning the effect of rootedness and regional and temporal differences are addressed.

The emergence and diffusion of fertility control strategies as part of the demographic transition are usually not occurring randomly in space and time. Next to individual-level characteristics also prevailing socio-economic contextual conditions as well as geographic characteristics such as distance to centers of the decline seem to be relevant. However, most existing studies on the fertility decline focus either on macro-level trends or on micro-level studies with limited geographic scope. With this paper we aim to contribute to close this existing research gap. We use 100% individual-level samples of the Swedish censuses in the years 1880, 1890 and 1900 which include detailed information on socio-economic status. Multi-level models are applied to link these individuals to contextual information on the local parishes they were living in. Our main research question is whether in this initial phase of the fertility decline in Sweden social distance or spatial distance were more relevant as constraints for the diffusion process. Our preliminary results suggest that in this initial phase of the decline social class differences were putting higher constraints on the diffusion of the fertility decline compared to spatial distances. This is in line with theoretical considerations by Szreter (1996) on “communication communities”.

Sub-national estimates of under-5 mortality are useful for evaluating within-country inequality, tracking progress, and identifying areas of greatest need. We estimate under-5 mortality for each of Zambia's 72 districts annually 1980-2010, using summary birth history data from censuses and complete birth history data from Demographic and Health surveys to fit a series of small area models. We consider a variety of generalized linear mixed models that differ in how spatial trends, temporal trends, and spatial-temporal interactions are introduced. All models suggest considerable heterogeneity in levels of under-5 mortality, with the worst off districts experiencing mortality risks 2-3 times as great as those in the best off districts. Distinct spatial trends are also apparent: districts in the northeast and southwest experience noticeably higher mortality than districts in the central part of the country. Progress in decreasing mortality over the past 30 years has also been variable: while there is some evidence of decline in most districts, our models suggest that a subset of districts have experienced decreases in mortality exceeding 50%.

It has been argued that social conditions are the fundamental determinants of health for those agents who possess them, and this argument has been bolstered by both individual and ecological studies. However, little is known about whether social conditions also benefit others nearby. Using the US county mortality data, we fill this gap by first theorizing the relationships between the social conditions of a county and the mortality of its neighbors with spatial spillover and social relativity perspectives. We then measured social conditions with income inequality, social capital, social affluence and concentrated disadvantage and used spatial Durbin modeling and spatial partitioning technique to examine the effects of these variables on mortality across space. The analytic results suggested that (1) social conditions of a specific county are not only related to its own mortality but also the mortality in neighboring counties; (2) The partitioning results provide evidence for spatial feedback, which underscores the importance of spatial structure underlying the data; and (3) The immediate neighbors (those shared the same boundaries or a vertex) play a more important role in understanding the direct impacts of social conditions on mortality than those neighbors far away. These findings provide new insight to mortality research.

For many countries, in particular in sub-Saharan Africa, Demographic and Health Surveys (DHS) are the main national source of data (depending on the subject). Several DHS collect latitude and longitude of surveyed clusters but the sampling method is not appropriate to derive local estimates: sample size is not large enough for a direct spatial interpolation.
We develop a methodological approach for estimating a proportion by using kernel density estimators with adaptive bandwidths of equal number of persons surveyed. The method was tested by creating a fictitious country from which survey datasets were produced. We compared the prevalence surface estimated from survey data with the model’s original prevalence surface.
This method makes it possible to achieve a smoothing effect that adapts to the high irregularity of spatial distribution among the survey clusters. The surfaces thus generated are relatively accurate for densely populated areas and strongly smoothed in sparsely surveyed areas. Although local variations were filtered out, the regional component in the spatial variation of prevalence was reproduced, and the estimated prevalence surfaces could be interpreted as regional trend surfaces.
Furthermore, this approach could be easily applied using prevR, a dedicated package for the statistical software R.

I argue that while demography is inherently a spatial science the training many demographers receive in fundamental spatial concepts, geospatial data and analytical methods is often limited, patchwork or nonexistent. However many important demographic questions deserve to be studied and framed using spatial approaches and this will become even more evident as changes in the volume, source, and form of available demographic data – much of it geocoded – further changes the data landscape and thus the methods demographers need to utilize. Ultimately changes in the data demographers use (how they collect, link, and analyze data) suggest the need to train next-generation population scientists in spatial thinking, concepts and methods of analysis. This will be a challenge given the existing logistical constraints on demography training. That is, any emergent method has to compete for its own place within the established curricula which will include core courses (e.g., fertility, migration, mortality), an expansive range of substantive cores (e.g., population and environment, urbanization) and other methods courses (e.g., demographic techniques, event history analysis, multilevel modeling). In this paper I propose and discuss some potential solutions to promoting and facilitating instruction in spatial demography across the globe.

Using geographically weighted regression (GWR), a recent study by Shoff and colleagues (2012) investigated the place-specific risk factors for prenatal care utilization in the U.S. and found that most of the relationships between late or not prenatal care and its determinants are spatially heterogeneous. However, the GWR approach may be subject to the confounding effect of spatial homogeneity. The goal of this study is to address this concern by including both spatial homogeneity and heterogeneity into the analysis. Specifically, we employ an analytic framework where a spatial lagged (SL) effect of the dependent variable is incorporated into the GWR model, which is called GWR-SL. Using this innovative framework, we found preliminary evidence to argue that spatial homogeneity may play a role in the study by Shoff et al. (2012). The GWR-SL approach allows us to gain a better understanding of prenatal care utilization in US counties, and improves on conventional approaches (e.g., OLS and spatial lag models). The new findings help us to better estimate how the predictors are associated with prenatal care utilization across space, and determine the counties whose neighboring prenatal care utilization matters.