| Preface | p. xv |
| Acknowledgements | p. xvii |
| Introduction | p. 1 |
| About the book | p. 1 |
| What is spatial data analysis? | p. 4 |
| Motivation for the book | p. 5 |
| Organization | p. 8 |
| The spatial data matrix | p. 10 |
| The context for spatial data analysis | |
| Spatial data analysis: scientific and policy context | p. 15 |
| Spatial data analysis in science | p. 15 |
| Generic issues of place, context and space in scientific explanation | p. 16 |
| Location as place and context | p. 16 |
| Location and spatial relationships | p. 18 |
| Spatial processes | p. 21 |
| Place and space in specific areas of scientific explanation | p. 22 |
| Defining spatial subdisciplines | p. 22 |
| Examples: selected research areas | p. 24 |
| Environmental criminology | p. 24 |
| Geographical and environmental (spatial) epidemiology | p. 26 |
| Regional economics and the new economic geography | p. 29 |
| Urban studies | p. 31 |
| Environmental sciences | p. 32 |
| Spatial data analysis in problem solving | p. 33 |
| Spatial data analysis in the policy area | p. 36 |
| Some examples of problems that arise in analysing spatial data | p. 40 |
| Description and map interpretation | p. 40 |
| Information redundancy | p. 41 |
| Modelling | p. 41 |
| Concluding remarks | p. 41 |
| The nature of spatial data | p. 43 |
| The spatial data matrix: conceptualization and representation issues | p. 44 |
| Geographic space: objects, fields and geometric representations | p. 44 |
| Geographic space: spatial dependence in attribute values | p. 46 |
| Variables | p. 47 |
| Classifying variables | p. 48 |
| Levels of measurement | p. 50 |
| Sample or population? | p. 51 |
| The spatial data matrix: its form | p. 54 |
| The spatial data matrix: its quality | p. 57 |
| Model quality | p. 58 |
| Attribute representation | p. 59 |
| Spatial representation: general considerations | p. 59 |
| Spatial representation: resolution and aggregation | p. 61 |
| Data quality | p. 61 |
| Accuracy | p. 63 |
| Resolution | p. 67 |
| Consistency | p. 70 |
| Completeness | p. 71 |
| Quantifying spatial dependence | p. 74 |
| Fields: data from two-dimensional continuous space | p. 74 |
| Objects: data from two-dimensional discrete space | p. 79 |
| Concluding remarks | p. 87 |
| Spatial data: obtaining data and quality issues | |
| Obtaining spatial data through sampling | p. 91 |
| Sources of spatial data | p. 91 |
| Spatial sampling | p. 93 |
| The purpose and conduct of spatial sampling | p. 93 |
| Design- and model-based approaches to spatial sampling | p. 96 |
| Design-based approach to sampling | p. 96 |
| Model-based approach to sampling | p. 98 |
| Comparative comments | p. 99 |
| Sampling plans | p. 100 |
| Selected sampling problems | p. 103 |
| Design-based estimation of the population mean | p. 103 |
| Model-based estimation of means | p. 106 |
| Spatial prediction | p. 107 |
| Sampling to identify extreme values or detect rare events | p. 108 |
| Maps through simulation | p. 113 |
| Data quality: implications for spatial data analysis | p. 116 |
| Errors in data and spatial data analysis | p. 116 |
| Models for measurement error | p. 116 |
| Independent error models | p. 117 |
| Spatially correlated error models | p. 118 |
| Gross errors | p. 119 |
| Distributional outliers | p. 119 |
| Spatial outliers | p. 122 |
| Testing for outliers in large data sets | p. 123 |
| Error propagation | p. 124 |
| Data resolution and spatial data analysis | p. 127 |
| Variable precision and tests of significance | p. 128 |
| The change of support problem | p. 129 |
| Change of support in geostatistics | p. 129 |
| Areal interpolation | p. 131 |
| Analysing relationships using aggregate data | p. 138 |
| Ecological inference: parameter estimation | p. 141 |
| Ecological inference in environmental epidemiology: identifying valid hypotheses | p. 147 |
| The modifiable areal units problem (MAUP) | p. 150 |
| Data consistency and spatial data analysis | p. 151 |
| Data completeness and spatial data analysis | p. 152 |
| The missing-data problem | p. 154 |
| Approaches to analysis when data are missing | p. 156 |
| Approaches to analysis when spatial data are missing | p. 159 |
| Spatial interpolation, spatial prediction | p. 164 |
| Boundaries, weights matrices and data completeness | p. 174 |
| Concluding remarks | p. 177 |
| The exploratory analysis of spatial data | |
| Exploratory spatial data analysis: conceptual models | p. 181 |
| EDA and ESDA | p. 181 |
| Conceptual models of spatial variation | p. 183 |
| The regional model | p. 183 |
| Spatial 'rough' and 'smooth' | p. 184 |
| Scales of spatial variation | p. 185 |
| Exploratory spatial data analysis: visualization methods | p. 188 |
| Data visualization and exploratory data analysis | p. 188 |
| Data visualization: approaches and tasks | p. 189 |
| Data visualization: developments through computers | p. 192 |
| Data visualization: selected techniques | p. 193 |
| Visualizing spatial data | p. 194 |
| Data preparation issues for aggregated data: variable values | p. 194 |
| Data preparation issues for aggregated data: the spatial framework | p. 199 |
| Non-spatial approaches to region building | p. 200 |
| Spatial approaches to region building | p. 201 |
| Design criteria for region building | p. 203 |
| Special issues in the visualization of spatial data | p. 206 |
| Data visualization and exploratory spatial data analysis | p. 210 |
| Spatial data visualization: selected techniques for univariate data | p. 211 |
| Methods for data associated with point or area objects | p. 211 |
| Methods for data from a continuous surface | p. 215 |
| Spatial data visualization: selected techniques for bi- and multi-variate data | p. 218 |
| Uptake of breast cancer screening in Sheffield | p. 219 |
| Concluding remarks | p. 225 |
| Exploratory spatial data analysis: numerical methods | p. 226 |
| Smoothing methods | p. 227 |
| Resistant smoothing of graph plots | p. 227 |
| Resistant description of spatial dependencies | p. 228 |
| Map smoothing | p. 228 |
| Simple mean and median smoothers | p. 230 |
| Introducing distance weighting | p. 230 |
| Smoothing rates | p. 232 |
| Non-linear smoothing: headbanging | p. 234 |
| Non-linear smoothing: median polishing | p. 236 |
| Some comparative examples | p. 237 |
| The exploratory identification of global map properties: overall clustering | p. 237 |
| Clustering in area data | p. 242 |
| Clustering in a marked point pattern | p. 247 |
| The exploratory identification of local map properties | p. 250 |
| Cluster detection | p. 251 |
| Area data | p. 251 |
| Inhomogeneous point data | p. 259 |
| Focused tests | p. 263 |
| Map comparison | p. 265 |
| Bivariate association | p. 265 |
| Spatial association | p. 268 |
| Hypothesis testing and spatial autocorrelation | |
| Hypothesis testing in the presence of spatial dependence | p. 273 |
| Spatial autocorrelation and testing the mean of a spatial data set | p. 275 |
| Spatial autocorrelation and tests of bivariate association | p. 278 |
| Pearson's product moment correlation coefficient | p. 278 |
| Chi-square tests for contingency tables | p. 283 |
| Modelling spatial data | |
| Models for the statistical analysis of spatial data | p. 289 |
| Descriptive models | p. 292 |
| Models for large-scale spatial variation | p. 293 |
| Models for small-scale spatial variation | p. 293 |
| Models for data from a surface | p. 293 |
| Models for continuous-valued area data | p. 297 |
| Models for discrete-valued area data | p. 304 |
| Models with several scales of spatial variation | p. 306 |
| Hierarchical Bayesian models | p. 307 |
| Explanatory models | p. 312 |
| Models for continuous-valued response variables: normal regression models | p. 312 |
| Models for discrete-valued area data: generalized linear models | p. 316 |
| Hierarchical models | |
| Adding covariates to hierarchical Bayesian models | p. 320 |
| Modelling spatial context: multi-level models | p. 321 |
| Statistical modelling of spatial variation: descriptive modelling | p. 325 |
| Models for representing spatial variation | p. 325 |
| Models for continuous-valued variables | p. 326 |
| Trend surface models with independent errors | p. 326 |
| Semi-variogram and covariance models | p. 327 |
| Trend surface models with spatially correlated errors | p. 331 |
| Models for discrete-valued variables | p. 334 |
| Some general problems in modelling spatial variation | p. 338 |
| Hierarchical Bayesian models | p. 339 |
| Statistical modelling of spatial variation: explanatory modelling | p. 350 |
| Methodologies for spatial data modelling | p. 350 |
| The 'classical' approach | p. 350 |
| The econometric approach | p. 353 |
| A general spatial specification | p. 355 |
| Two models of spatial pricing | p. 356 |
| A 'data-driven' methodology | p. 358 |
| Some applications of linear modelling of spatial data | p. 358 |
| Testing for regional income convergence | p. 359 |
| Models for binary responses | p. 361 |
| A logistic model with spatial lags on the covariates | p. 361 |
| Autologistic models with covariates | p. 364 |
| Multi-level modelling | p. 365 |
| Bayesian modelling of burglaries in Sheffield | p. 367 |
| Bayesian modelling of children excluded from school | p. 376 |
| Concluding comments | p. 378 |
| Software | p. 379 |
| Cambridgeshire lung cancer data | p. 381 |
| Sheffield burglary data | p. 385 |
| Children excluded from school: Sheffield | p. 391 |
| References | p. 394 |
| Index | p. 424 |
| Table of Contents provided by Ingram. All Rights Reserved. |
