Article

Temporal transitions of demographic dot maps

Authors:
To read the full-text of this research, you can request a copy directly from the author.

Abstract

Dot maps are often used to display the distributions of populations over space. This paper details a method for extending dot maps in order to visualize changes in spatial patterns over time. Specifically, we outline a selective linear interpolation procedure to encode the time range in which dots are visible on a map, which then allows for temporal queries and animation. This methodology is exemplified first by animating population growth across the United States, and second, through an interactive application showing changing poverty distributions in Toronto, Canada.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the author.

ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Article
Although the effects of urban shrinkage on quality of life and the built environment have received a great deal of attention, the characteristics of those experiencing these impacts have been much less studied. This is ironic, as urban shrinkage or depopulation is by nature a demographic phenomenon: city sizes evolve precisely because people move in and out, are born, and die. Moreover, the demographic processes that contribute to shrinking cities—out-migration and death—are selective and so they also govern who remains behind in cities as they shrink. It is this latter group that is the focus of this research. The analysis contributes to the literature on shrinking cities through its novel consideration of community-level exposure to depopulation. In particular, it investigates who is impacted by loss; the extent to which population loss is experienced disproportionately across urban space and demographic subgroups; and whether decline occurring at multiple spatial scales magnifies exposure for some groups more than others. Findings show that, at both city and census tract levels, demographic characteristics of growth and loss areas are different and, at all levels, some groups are more likely than others to be living in a loss-impacted area.
Full-text available
Article
Racial geography, mapping spatial distributions of different racial groups, is of keen interest in a multiracial society like the United States. A racial dot map is a method of visualizing racial geography, which depicts spatial distribution, population density, and racial mix in a single, easy-to-understand map. Because of the richness of information it carries, the dot map is an excellent tool for visual analysis of racial distribution. Presently-used racial dot maps are based on the Census data at the tract or the block level. In this paper, we present a method of constructing a more spatially-accurate racial dot map based on a sub-block-resolution population grid. The utility of our dot maps is further enhanced by placing dots on the map in random order regardless of the race they represent in order to achieve a more accurate depiction of local racial composition. We present a series of comparisons between dot maps based on tract, block, and grid data. The advantage of a grid-based dot map is evident from the visual comparison of all maps with an actual image of the mapped area. We make available the R code for constructing grid-based dot maps. We also make available 2010 grid-based racial dot maps for all counties in the conterminous United States.
Full-text available
Article
Dot mapping is a traditional method for visualizing quantitative data, but current automated dot mapping techniques are limited. The most common automated method places dots pseudo-randomly within enumeration areas, which can result in overlapping dots and very dense dot clusters for areas with large values. These issues affect users’ ability to estimate values. Graduated dot maps use dots with different sizes that represent different values. With graduated dot maps the number of dots on a map is smaller, reducing the likelihood of overlapping dots. This research introduces an automated method of generating graduated dot maps that arranges dots with blue-noise patterns to avoid overlap and uses clustering algorithms to replace densely packed dots with those of larger sizes. A user study comparing graduated dot maps, pseudo-random dot maps, blue-noise dot maps, and proportional circle maps with almost 300 participants was conducted. Results indicate that map users can more accurately extract values from graduated dot maps than from the other map types. This is likely due to the smaller number of dots per enumeration area in graduated dot maps. Map users also appear to prefer graduated dot maps over other map types.
Full-text available
Article
Geographical mapping of infectious diseases is an important tool for detecting and characterising outbreaks. Two common mapping methods, dot maps and incidence maps, have important shortcomings. The former does not represent population density and can compromise case privacy, and the latter relies on pre-defined administrative boundaries. We propose a method that overcomes these limitations: dot map cartograms. These create a point pattern of cases while reshaping spatial units, such that spatial area becomes proportional to population size. We compared these dot map cartograms with standard dot maps and incidence maps on four criteria, using two example datasets. Dot map cartograms were able to illustrate both incidence and absolute numbers of cases (criterion 1): they revealed potential source locations (Q fever, the Netherlands) and clusters with high incidence (pertussis, Germany). Unlike incidence maps, they were insensitive to choices regarding spatial scale (criterion 2). Dot map cartograms ensured the privacy of cases (criterion 3) by spatial distortion; however, this occurred at the expense of recognition of locations (criterion 4). We demonstrate that dot map cartograms are a valuable method for detection and visualisation of infectious disease outbreaks, which facilitates informed and appropriate actions by public health professionals, to investigate and control outbreaks. © 2017, European Centre for Disease Prevention and Control (ECDC). All rights reserved.
Full-text available
Article
Interactive visual applications often rely on animation to transition from one display state to another. There are multiple animation techniques to choose from, and it is not always clear which should produce the best visual correspondences between display elements. One major factor is whether the animation relies on staggering-an incremental delay in start times across the moving elements. It has been suggested that staggering may reduce occlusion, while also reducing display complexity and producing less overwhelming animations, though no empirical evidence has demonstrated these advantages. Work in perceptual psychology does show that reducing occlusion, and reducing inter-object proximity (crowding) more generally, improves performance in multiple object tracking. We ran simulations confirming that staggering can in some cases reduce crowding in animated transitions involving dot clouds (as found in, e.g., animated 2D scatterplots). We empirically evaluated the effect of two staggering techniques on tracking tasks, focusing on cases that should most favour staggering. We found that introducing staggering has a negligible, or even negative, impact on multiple object tracking performance. The potential benefits of staggering may be outweighed by strong costs: a loss of common-motion grouping information about which objects travel in similar paths, and less predictability about when any specific object would begin to move. Staggering may be beneficial in some conditions, but they have yet to be demonstrated. The present results are a significant step toward a better understanding of animation pacing, and provide direction for further research.
Full-text available
Article
Dot mapping is a cartographic representation method to visualise discrete absolute values and their spatial distribution. To achieve this, dots equal in size and represented value are used. According to the dot value, a certain number of dots are used to depict a data value. These dots usually form dot clusters. The data value needs to be rounded to a multiple of the dot value. It is possible to roughly determine the visualised data value by counting the dots and multiplying this number with the dot value. As there are many parameters – dot size, dot value, map scale – to consider when designing a dot map, the manual way is very complex and time consuming. This paper presents a method to automatically create a dot representation of a dot map from given statistical data that needs no cartographic expertise. The dot representation may be combined with other elements, such as a topographic background, to form a complete map. So the algorithm can easily be integrated into the map design process. The paper refines the basic approach of automated dot mapping published earlier. The dot placement and arrangement have been improved compared to the basic method.
Full-text available
Article
Differences in the reporting units of data from diverse sources and changes in units over time are common obstacles to analysis of areal data. We compare common approaches to this problem in the context of changes over time in the boundaries of U.S. census tracts. In every decennial census, many tracts are split, consolidated, or changed in other ways from the previous boundaries to reflect population growth or decline. We examine two interpolation methods to create a bridge between years, one that relies only on areal weighting and another that also introduces population weights. Results demonstrate that these approaches produce substantially different estimates for variables that involve population counts, but they have a high degree of convergence for variables defined as rates or averages. Finally, the article describes the Longitudinal Tract Database (LTDB), through which we are making available public-use tools to implement these methods to create estimates within 2010 tract boundaries for any tract-level data (from the census or other sources) that are available for prior years as early as 1970.
Full-text available
Article
The Canadian Century Research Infrastructure (CCRI) aims to build historical data sets from the 1911–1951 Canadian census manuscripts. To enhance the meaning of the microdata, and to enable users to apprehend the spatial dimension of the phenomena they study, the CCRI integrates as much geographical information as the census can provide and also provides users with resources to map selected data or the results they get. A framework to conduct basic or sophisticated spatial analyses is also provided. The integration of geographical information necessitates the reconstitution of the entire geography of census taking, as well as of census dissemination, through the first half of the twentieth century. The raw material of the census manuscript is organized according to enumeration areas, whereas the basic spatial unit of dissemination is the census subdivision listed in the published aggregate census returns. To spatially process and map the microdata as entered and the aggregate data as published, polygon files are being created for nearly 32,000 subdivisions and managed within a geodatabase. The files form the fundamental fabric to which microdata and aggregate data are both linked. The authors address the unexpected challenges that arise from the realities of historical census data and provide an overview of the limitations and the technical framework of the overall geographical component of the CCRI.
Full-text available
Article
In this paper dot maps are used as an epidemiological tool to describe the spatial patterns of cases, non-cases and some relevant risk factors associated with the occurrence of Schistosoma mansoni infection in an urban environment. The study was carried out in Santo Antonio de Jesus, a town in north-east Brazil with a population of around 45000 inhabitants. The survey was targeted at all children aged between 12 years 8 months and 14 years 11 months who were living in the town at the time of the survey. An extensive questionnaire was used to collect information on each child and on family and household conditions; stod samples were also taken for quantitative examinetion. A computerized digitizer was used on a map (scale 1:4000) to create x and y coordinates for each child's household. The result was matched with individual data from the questionnaires and the stool examination. From this data base it was pussible, by using a statistical package, to make epidemiological analyses and by using the statistical package associated with a geocartographic system package it was possible to dot plot selected values of defined variables. These techniques substantiated the findings on the risk factors at a local level and, as a consequence, improve our ability to define targeted actions towards local control of schistwomiasis.
Article
Many cities have undergone spatial redistributions of low-income populations from central to suburban neighborhoods over the past several decades. A potential negative impact of these trends is that low-income populations are concentrating in more automobile-oriented areas, resulting in increased barriers to daily travel and activity participation, particularly for those who are unable to afford a private vehicle. Accordingly, the objective of this article is to analyze the links between increasing sociospatial inequalities, transport disadvantage, and adverse travel behavior outcomes. This is examined first from a theoretical perspective and second via a spatiotemporal analysis for the Toronto region from 1991 to 2016. Findings show that many suburban areas in Toronto are not only declining in socioeconomic status but are also suffering from increased barriers to daily travel evidenced by longer commute times and decreasing activity participation rates, relative to central neighborhoods. Because of these adverse effects, this evidence further supports the need for progressive planning and policy aimed at curbing continuing trends of suburbanization of poverty while also improving levels of transport accessibility in the suburbs.
Article
Dot maps are effective for cartographic visualization of categorical data. Recent advances in Web mapping technology have facilitated the development of interactive dot maps, in which users can pan and zoom to view data distributions for different areas. This interactivity, however, introduces multiple cartographic challenges, as design decisions that are appropriate at large scales can lead to clutter and illegibility at small scales. This article considers these challenges in the context of an applied example - an interactive dot map of educational attainment in the United States. It covers the methodology of the map's creation as well as how it addresses the cartographic challenges of interactive dot mapping.
Article
en Performing longitudinal analysis of socio‐economic change in small‐area spatial units such as census tracts presents several methodological complications and requires significant data preparation. Unit boundaries are revised each census year because of changes in population and delineation methodologies. This limits cross‐year comparison since data are not representative of the same spatial units. To address these problems, we have developed an innovative procedure to reduce error when comparing tract‐level data across census years by apportioning data to the same areal units. This paper describes the methods used to create the Canadian Longitudinal Tract Database. Our procedure is a combination of map‐matching techniques, dasymetric overlays, and population‐weighted areal interpolation. The output is a set of tables with apportionment weights pertaining to pairs of unique boundary identifiers across census years, which can be linked with census data or other data with census identifiers that require longitudinal comparison. Un nouvel outil pour la recherche sur les changements dans le voisinage : la base de données longitudinales canadiennes des secteurs de recensement, 1971–2016 fr La réalisation d'une analyse longitudinale des changements socioéconomiques dans des unités spatiales de zone restreinte, par exemple des secteurs de recensement, présente plusieurs complications méthodologiques et requiert une préparation importante de données. Les limites des unités sont révisées chaque année de recensement en raison des changements dans la population et dans les méthodologies de délimitation. Ceci limite la comparaison des différentes années puisque les données ne sont pas représentatives des mêmes unités spatiales. Pour résoudre ces problèmes, nous avons développé une procédure innovatrice pour réduire les erreurs lors de la comparaison des données au niveau des secteurs entre les années de recensement en affectant les données aux mêmes unités spatiales. La présente communication décrit les méthodes utilisées pour créer la base de données longitudinales canadiennes des secteurs de recensement. Notre procédure est une combinaison de techniques de corrélation de cartes, de superpositions dasymétriques et d'interpolation spatiale pondérée selon la population. Le résultat est une série de tableaux avec une répartition des pondérations se rapportant à des paires d'identificateurs uniques des limites pour chacune des années de recensement qui peuvent être reliées aux données du recensement ou à d'autres données avec des identificateurs du recensement qui requièrent une comparaison longitudinale.
Article
Unit visualizations are a family of visualizations where every data item is represented by a unique visual mark—a visual unit—during visual encoding. For certain datasets and tasks, unit visualizations can provide more information, better match the user's mental model, and enable novel interactions compared to traditional aggregated visualizations. Current visualization grammars cannot fully describe the unit visualization family. In this paper, we characterize the design space of unit visualizations to derive a grammar that can express them. The resulting grammar is called ATOM, and is based on passing data through a series of layout operations that divide the output of previous operations recursively until the size and position of every data point can be determined. We evaluate the expressive power of the grammar by both using it to describe existing unit visualizations, as well as to suggest new unit visualizations.
Article
Rapid transformations sweeping the United States over the past 50 years have necessitated a reassessment of longstanding theories on how the neighborhood change process has unfolded. This article builds upon recent methodological advancements aimed at understanding longitudinal dynamics by developing a workflow that blends the self-organizing map and a sequential alignment method to visualize pathways of change in a multivariate context. It identifies the predominant pathways in which neighborhoods have changed according to their racial, ethnic, socioeconomic and housing characteristics in the largest US metropolitan statistical areas from 1980 to 2010. The distribution of these pathways is subsequently examined between metropolitan statistical areas and the spatial clustering of these trajectories within cities is analyzed. Results reveal a white-flight type process, the establishment of a multiethnic neighborhood, densification of single-family neighborhoods, gentrification in relatively diverse neighborhoods, upgrading of white single family neighborhoods, and the most frequent pathway of all: no change. High-poverty minority and wealthy white neighborhoods are most spatially compact and expanding in a contiguous manner, while multiethnic neighborhoods are relatively dispersed. Six groups of metropolitan statistical areas are identified based upon the similarity of their neighborhood composition. Parallels are drawn between the formation of enduring high-poverty black neighborhoods in Northern and Midwestern cities and the emergence of clusters high-poverty Hispanic neighborhoods in Hispanic destination cities.
Article
High resolution colour mapping using modern technology has allowed us to explore the breadth of census data available to the contemporary social scientist. An examination of methods used to create cartograms, which minimise visual bias involves considering how densities and area boundaries should be mapped. The Modifiable Area Unit Problem is discussed and the advantages of using three- colour or trivariate mapping are outlined. Further modifications to traditional census mapping such as the mapping of change and flow are addressed.
Article
Dot maps are one of the best ways to visualize absolute values in thematic cartography. Dots represent quantitative data on a map. Population is often used in this type of representation. This paper presents a population dot density map for the year 2011 on two scales: (1) for mainland Portugal, and (2) for the Lisbon and Oporto regions. We have used dots with constant values and sizes at the most detailed statistical level (i.e. statistical subsection) for localities with less than 5000 inhabitants, and proportional circles for localities with more than 5000 inhabitants. These two scales of analysis coupled with two cartographic representation techniques used on a single map allow for a clear reading of the distribution of population.
Article
Paying the tribute to the 100 years anniversary of dasymetric mapping, this paper aims to provide a detailed inquiry into historical beginnings of this cartographic technique by working with literature and archival sources, primarily with the writings of the method’s inventor, Benjamin Semenov-Tian-Shansky. The paper confirms that the dasymetric method was first proposed in 1911 and its first cartographic products appeared in 1919. The paper reviews and discusses the dasymetric technique used by Semenov-Tian-Shansky in the 1920s to produce the ‘Dasymetric Map of European Russia’ that, perhaps, still remains one of the most extensive dasymetric mapping projects in the history of cartography. This research will help to continue reconstructing the history of dasymetric mapping and will aid in better understanding of the original technique itself and its relations with later achievements in cartography and GIScience.
Article
Geographic data include locations of places and events, and spatial changes of phenomena. Time is an essential element in tracking changes and in analyzing them to establish patterns, to predict future events, or to explain occurrences. Time information, however, comes in different forms when geographic data are collected. It may be a simple time stamp: the date when a city was incorporated. It may be an expression of duration: how long the forest fires burned. Geographic data may be collected at regular intervals, e.g., the decennial census in the USA. Or they may be recorded only when a change occurs, e.g. a hillside is cut away for development and the change is recorded on the topographic map. Time may be used as a distance measure, and space may be used as a delineator of time. Because of these variations in the uses of time, various types of time information must be understood and the understanding incorporated into methods of producing maps so as to show the information clearly and efficiently. This research categorizes time as it is used by geographers and as it appears in different forms on maps. The categories are Moments, Duration, Structured Time, Time as Distance, and Space as Clock. Each of these categories defines a particular use of time and provides the basic symbology that is used to represent spatiotemporal information on maps according to the spatial dimensions (point, line, or area) typically available to cartographers. The framework for the symbolization of temporal information on maps presented in this research establishes two things: the various categories of time found in geography; and the spatial dimensions and symbology available to the cartographer for representing spatiotemporal information on maps.
Article
Recent studies on urban poverty in Canadian cities suggest a growing spatial concentration of poor populations within metropolitan regions. This article assesses trends in the intra-urban distribution of the poor population from 1986 to 2006 in eight of Canada's largest cities. We consider five well-known dimensions of segregation, as identified by, in order to examine changes in the spatial distribution of poor populations within metropolitan areas: evenness, exposure, concentration, clustering, and centralization. These indices were calculated for low-income populations at the census tract level using data from five Canadian censuses. Although each metropolitan area has distinctive characteristics, we were able to identify some general trends. The results suggest that, in 2006 compared to 1986, low-income populations lived in more spatially concentrated areas, which were, at the same time, socioeconomically more homogeneous and more dispersed throughout the metropolitan area. In addition, we observed that over the last twenty years areas of poverty have been located, for the most part, in neighbourhoods adjacent to downtown cores. Nevertheless, we found that poverty has mostly increased in suburban areas located outside inner-city neighbourhoods. Growing socioeconomic homogeneity and dispersion of low income areas in metropolitan areas reveal new spatial patterns of urban poverty distribution. These findings should be cause for concern as social isolation in the most disadvantaged neighbourhoods could affect the life chances and opportunities for the residents of those areas.
Article
Dot mapping is a cartographic representation method to visualize quantitative absolute values, e.g. population figures, and spatial distribution patterns. In the course of the development of computer-assisted cartography many cartographic methods have already been implemented in computer algorithms, but for dot mapping only a few solutions with some considerable constraints have been found until now. The most important constraint is the lack of repeatable results because these solutions work with random dot placement. This may result in different maps, where the only real difference is the date of production, not the data that is mapped. Another constraint is the overlap of dots which may interfere with the user's intention to count the dots. Although, this last constraint may not be considered important by all, the first constraint is clearly problematic for the cartographer. This paper deals with the automated production of dot distribution maps which will not suffer from the above named constraints. The paper focuses on the placement of dots. It presents a method of placing dots in a centred manner as dot clusters grouping around one central point, e.g. the centroid of an enumeration area. The geometric basis is a spiral whirl. Besides the regular basis, the dot clusters should not be perceived as strict geometric figures and a pseudo-random element is therefore introduced to blur the symmetry of the dot clusters. Although, this blurring is apparently random, it is repeatable. The new approach is a first step to produce dot maps automatically by providing a mathematical structure to connect the area needed for placing the dots with the number and size of the dots. The method is meant to support cartographic laypersons as well as professional cartographers in the process of map design.
Article
This study compares the changes in the household characteristics and income distributions of men and women in the Toronto census tracts that gentrified in the 1970s with the changes taking place across the metropolitan area. Higher-income women were gentrifiers from the beginning and lower-income men were more likely to be displaced. The factors driving gentrification are also transforming the rest of the metropolitan area, making the continuation of gentrification appear inevitable unless aggressive social housing programs are implemented to secure tenure for lower-income households and help mitigate the loss of lower priced housing in the inner city.
Article
This paper describes techniques to compute and map dasymetric population densities and to areally interpolate census data using dasymetrically derived population weights. These techniques are demonstrated with 1980-2000 census data from the 13-county Atlanta metropolitan area. Land-use/land-cover data derived from remotely sensed satellite imagery were used to determine the areal extent of populated areas, which in turn served as the denominator for dasymetric population density computations at the census tract level. The dasymetric method accounts for the spatial distribution of population within administrative areas, yielding more precise population density estimates than the choroplethic method, while graphically representing the geographic distribution of populations. In order to areally interpolate census data from one set of census tract boundaries to another, the percentages of populated areas affected by boundary changes in each affected tract were used as adjustment weights for census data at the census tract level, where census tract boundary shifts made temporal data comparisons difficult. This method of areal interpolation made it possible to represent three years of census data (1980, 1990, and 2000) in one set of common census tracts (1990). Accuracy assessment of the dasymetrically derived adjustment weights indicated a satisfactory level of accuracy. Dasymetrically derived areal interpolation weights can be applied to any type of geographic boundary re-aggregation, such as from census tracts to zip code tabulation areas, from census tracts to local school districts, from zip code areas to telephone exchange prefix areas, and for electoral redistricting.
Article
Six distinct generalizations useful in dynamic cartography demonstrate the importance of the temporal dimension in map generalization.
Article
Dot maps show the geographic distribution of a phenomenon in an area by placing dots representing a certain quantity of the phenomenon where it is most likely to occur. The fundamental steps in dot mapping are selecting the dot size, determining the dot unit value, and placing the correct number of dots in a manner that correctly reflects the geographic distribution of features. Selecting the dot size is a subjective decision, but the dot unit value has long been determined with the aid of the Mackay nomograph. Close examination of the nomograph finds it is not appropriate for determining the dot unit value when dot placement is based on computer-generated random numbers that result in overlapping dots. A new graphical aid for the determination of dot unit value was created by modeling aggregate area of dots and amount of dot overlap using a truncated form of the unification equation from probability theory. Aggregate dot areas predicted by this equation were tested against actual random dots created for several common dot sizes, and high agreement was found between measured and predicted aggregate areas. Pseudo-random dot placement with a maximum overlap constraint for dot pairs appears to mimic more closely how cartographers have traditionally placed dots. Pseudo-random dot placement can be thought of as similar to rigid random placement of circles in a square with maximum circle overlap limits from mutually exclusive to completely random dots. Thinking of dot placement in this manner allowed a general equation for aggregate dot area to be devised as a weighted combination of the mutually exclusive and completely random dot endpoint equations. Aggregate areas predicted by this general equation were found to match actual assemblages of pseudo-random dots with differing maximum dot pair overlaps closely.
Article
Predictions of the 1960s about the computer's potential to change cartography are finally being fulfilled. Dynamic maps for vehicle navigation, interactive cartographic/statistical tools, and map animation are being investigated actively. As these new environments for mapping become available, we must reevaluate past questions about transformations from reality to data and data to map. In this paper, we consider these transformation questions in the context of statistical map animation. The issues discussed were raised in producing a “map movie” depicting the spread of acquired immune deficiency syndrome (AIDS) over time. Jenks' data model concept is used as the basis for a typology of data models representing phenomena typically depicted by enumeration unit data. The typology is then used to evaluate symbolization decisions for AIDS incidence maps. Implications for symbol selection imposed by dynamic rather than static maps are considered, as are technical issues involved in producing the animation on a microcomputer platform. A hybrid symbolization method that we have termed the “chorodot” is suggested as a way to meet the constraints on symbolization imposed by animation and to represent the appropriate data model for AIDS incidence.
Article
Creating dot maps to show changes in racial and Hispanic population distributions between two census periods can be an effective way to examine one of the most important dimensions of change within any metropolitan area. Using dots of one color to show population increase and dots of a second color to show population decrease vividly reveals where changes have occurred within a larger total population. We prepared such maps for the book Changing Faces, Changing Places: Mapping Southern Californians, the text of which analyzes and interprets the population shifts evident on the maps. The maps show the expansion and contraction of racial and Hispanic populations in specific neighborhoods so that community leaders and residents alike can easily relate general trends to their localities. In this article we describe the preparation of these dot maps and explain major problems encountered in linking the 1990 and 2000 census population counts at the tract level. We explain our solutions, which we believe made possible more accurate mapping of neighborhood change.
Article
High resolution colour mapping using modern technology has allowed us to explore the breadth of census data available to the contemporary social scientist. An examination of methods used to create cartograms, which minimise visual bias involves considering how densities and area boundaries should be mapped. The Modifiable Area Unit Problem is discussed and the advantages of using three colour or trivariate mapping are outlined. Further modifications to traditional census mapping such as the mapping of change and flow are addressed.
Article
Demographic forecasting techniques do not perform particularly well for small areas. In this study we propose a spatio-temporal regression approach for small-area population forecasting that borrows strength from what has happened nearby and what has happened in the past. In particular, a regression model incorporating temporally lagged neighbour growth and neighbour characteristics is applied to examine population change at the minor civil division (MCD) governmental level in Wisconsin, USA since 1960. For each MCD, the population growth rate for 1980–1990 is regressed on its growth rate for 1970–1980, its various characteristics in 1980, neighbour growth rates for 1970–1980, and neighbour characteristics in 1980. The estimated regression coefficients are then used for projecting population in 2000. Accuracy of the forecasts is measured against the 2000 Census counts. The state's official MCD projections for 2000, which were based on an extrapolation projection – the most often used traditional population forecasting approach for small geographic areas – are taken as the ‘gold standard’ against which improvements through the regression formulations are sought. The projection evaluations reveal mixed results and do not suggest unambiguous preference for the spatio-temporal regression approach or the extrapolation projection. We discuss several reasons for the inability of our spatio-temporal forecasting model to outshine the simple extrapolation forecast. Although this is disappointing, the proposed approach is more solidly theoretically grounded and provides useful information to policy and decision makers at the community level regarding the consequences of various developmental strategies being adopted by themselves as well as by their neighbours. Copyright © 2010 John Wiley & Sons, Ltd.
Article
Conducting temporal analysis of census data often requires applying areal interpolation to integrate data that have been spatially aggregated using incompatible zoning systems. This article introduces a method of areal interpolation, target-density weighting (TDW), that is useful for long-term temporal analysis because it requires only readily available historical data and basic geographic information system operations. Then, through regression analysis of a large sample of U.S. census tract data, a model is produced that relates the error in TDW estimates of tract population to four basic properties of tracts. An analysis of model residuals combined with theorized absolute limits on interpolation error yields formulas with which we can compute upper and lower prediction bounds on the population in a tract of one census at the time of a different census. These prediction intervals enable the interpretation of different interpolated estimates with appropriately varying degrees of uncertainty.
Article
This overview introduces the key structure of the field of information visualization, a number of influential exemplars in the field, and challenging as well as promising directions of future developments. The focus is on explaining some of the most fundamental concepts, prominent approaches, and commonly held criteria. The overview also aims to point out theoretical and practical challenges that the community as a whole has been addressing. Copyright © 2010 John Wiley & Sons, Inc. For further resources related to this article, please visit the WIREs website.
Article
Two experiments observed performance on a cluster identification task across a variety of common statistical maps. Stimulus maps displayed mortality rates for several diseases and subjects had to identify regions of the map that were perceived to form a cluster of particularly high (or low) mortality. Subjects marked the perceived centroid of each cluster, and analyses focused on the dispersion of centroid location across subjects. Under these circumstances, monochrome classed choropleth maps were found to minimize dispersion, compared to a two opposing colours scheme, a dot density map, a pie map, and a categorical (hue-based) colour scheme. Maps using a familiar geographical unit (i. e. a U. S. state) supported better recall of the information than maps using less familiar and smaller geographical units. The results were found to be interpretable within current cognitive theory.
Article
Since population censuses are not annually implemented, population estimates are needed for the intercensal period. This paper describes simultaneous implementations of the temporal interpolation and forecasting of the population census data, aggregated by age and period. Since age equals period minus cohort, age-period-cohort decomposition suffers from the identification problem. In order to overcome this problem, the Bayesian cohort (BC) model is applied. The efficacy of the BC model for temporal interpolation is examined in comparison with official Japanese population estimates. Empirical results suggest that the BC model is expected to work well in temporal interpolation. Regarding the age-period-cohort decomposition of the Japanese census data, it is shown that the cohort effect is the largest while the other two effects are very small but not negligible. With regard to the forecasting of the Japanese population, the official population forecast considerably outperforms the BC forecast in most forecast horizons. However, the pace of increase in root mean square error for longer-term forecasting is larger in the official population forecast than in the BC forecasts. As a result, a variant of the BC forecast is best for 10-year forecast. KeywordsAge-period-cohort decomposition-Cohort data-Forecasting-Interpolation-Population census
Carte philosophique figurant la population de la France
  • A J Frere De Montizon
England and Wales: distribution of the population. Census of Great Britain, 1851, Population Tables I. Her Majesty's Stationery Office
  • A Petermann
Cecrops, king of Athens: The first (?) recorded population census in history
  • S Missiakoulis
Historical Population Estimates for 2010 U.S. States, Counties and Metro/Micro Areas, 1790-2010. Data Repository for the University of Minnesota
  • J P Schroeder
Cartography: Thematic map design. Wm. C. Brown Publishers
  • B Dent
  • J Torguson
  • T Hodler
Dotting the dot map: An analysis of dot size, number, and visual tone density. Surveying and Mapping
  • R J Mackay
Income inequality and polarization in Canada's cities: An examination and new form of measurement
  • A Walks
Mapping time. Cartographica: The International Journal for Geographic Information and Geovisualization
  • I Vasiliev
The three cities within Toronto
  • J D Hulchanski