Spatial data analysis: theory and practice

Half-title......Page 3
Title......Page 5
Copyright......Page 6
Dedication......Page 7
Contents......Page 9
Preface......Page 17
Readership......Page 18
Acknowledgements......Page 19
Copyright acknowledgements......Page 21
0.1 About the book......Page 23
0.2 What is spatial data analysis?......Page 26
0.3 Motivation for the book......Page 27
0.4 Organization......Page 30
0.5 The spatial data matrix......Page 32
Part A The context for spatial data analysis......Page 35
1.1 Spatial data analysis in science......Page 37
(a) Location as place and context......Page 38
(b) Location and spatial relationships......Page 40
1.1.2 Spatial processes......Page 43
1.2.1 Defining spatial subdisciplines......Page 44
(a) Environmental criminology......Page 46
(b) Geographical and environmental (spatial) epidemiology......Page 48
(c) Regional economics and the new economic geography......Page 51
(d) Urban studies......Page 53
(e) Environmental sciences......Page 54
1.2.3 Spatial data analysis in problem solving......Page 55
1.3 Spatial data analysis in the policy area......Page 58
1.4.1 Description and map interpretation......Page 62
1.5 Concluding remarks......Page 63
2 The nature of spatial data......Page 65
2.1.1 Geographic space: objects, fields and geometric representations......Page 66
2.1.2 Geographic space: spatial dependence in attribute values......Page 68
2.1.3 Variables......Page 69
(a) Classifying variables......Page 70
(b) Levels of measurement......Page 72
2.1.4 Sample or population?......Page 73
2.2 The spatial data matrix: its form......Page 76
2.3 The spatial data matrix: its quality......Page 79
2.3.1 Model quality......Page 80
(b) Spatial representation: general considerations......Page 81
2.3.2 Data quality......Page 83
(a) Accuracy......Page 85
(b) Resolution......Page 89
(c) Consistency......Page 92
(d) Completeness......Page 93
(a) Fields: data from two-dimensional continuous space......Page 96
(b) Objects: data from two-dimensional discrete space......Page 101
2 .5 Concluding remarks......Page 109
Part B Spatial data: obtaining data and quality issues......Page 111
3.1 Sources of spatial data......Page 113
3.2.1 The purpose and conduct of spatial sampling......Page 115
(a) Design-based approach to sampling......Page 118
(b) Model-based approach to sampling......Page 120
(c) Comparative comments......Page 121
3.2.3 Sampling plans......Page 122
(a) Design-based estimation of the population mean......Page 125
(b) Model-based estimation of means......Page 128
(c) Spatial prediction......Page 129
(d) Sampling to identify extreme values or detect rare events......Page 130
3.3 Maps through simulation......Page 135
4.1.1 Models for measurement error......Page 138
(a) Independent error models......Page 139
(b) Spatially correlated error models......Page 140
(a) Distributional outliers......Page 141
(b) Spatial outliers......Page 144
(c) Testing for outliers in large data sets......Page 145
4.1.3 Error propagation......Page 146
4.2 Data resolution and spatial data analysis......Page 149
4.2.1 Variable precision and tests of significance......Page 150
(a) Change of support in geostatistics......Page 151
(b) Areal interpolation......Page 153
4.2.3 Analysing relationships using aggregate data......Page 160
(a) Ecological inference: parameter estimation......Page 163
(b) Ecological inference in environmental epidemiology: identifying valid hypotheses......Page 169
(c) The modifiable areal units problem (MAUP)......Page 172
4.3 Data consistency and spatial data analysis......Page 173
4.4 Data completeness and spatial data analysis......Page 174
4.4.1 The missing-data problem......Page 176
(a) Approaches to analysis when data are missing......Page 178
(b) Approaches to analysis when spatial data are missing......Page 181
4.4.2 Spatial interpolation, spatial prediction......Page 186
4.4.3 Boundaries, weights matrices and data completeness......Page 196
4.5 Concluding remarks......Page 199
Part C The exploratory analysis of spatial data......Page 201
5.1 EDA and ESDA......Page 203
(a) The regional model......Page 205
(b) Spatial ‘rough’ and ‘smooth’......Page 206
(c) Scales of spatial variation......Page 207
6.1 Data visualization and exploratory data analysis......Page 210
6.1.1 Data visualization: approaches and tasks......Page 211
6.1.2 Data visualization: developments through computers......Page 214
6.1.3 Data visualization: selected techniques......Page 215
6.2.1 Data preparation issues for aggregated data: variable values......Page 216
6.2.2 Data preparation issues for aggregated data: the spatial framework......Page 221
(a) Non-spatial approaches to region building......Page 222
(b) Spatial approaches to region building......Page 223
(c) Design criteria for region building......Page 225
6.2.3 Special issues in the visualization of spatial data......Page 228
6.3 Data visualization and exploratory spatial data analysis......Page 232
(a) Methods for data associated with point or area objects......Page 233
(b) Methods for data from a continuous surface......Page 237
6.3.2 Spatial data visualization: selected techniques for bi- and multi-variate data......Page 240
6.3.3 Uptake of breast cancer screening in Sheffield......Page 241
6.4 Concluding remarks......Page 247
7 Exploratory spatial data analysis: numerical methods......Page 248
7.1.1 Resistant smoothing of graph plots......Page 249
7.1.3 Map smoothing......Page 250
(b) Introducing distance weighting......Page 252
(c) Smoothing rates......Page 254
(d) Non-linear smoothing: headbanging......Page 256
(e) Non-linear smoothing: median polishing......Page 258
7.2 The exploratory identification of global map properties: overall clustering......Page 259
7.2.1 Clustering in area data......Page 264
7.2.2 Clustering in a marked point pattern......Page 269
7.3 The exploratory identification of local map properties......Page 272
(a) Area data......Page 273
(b) Inhomogeneous point data......Page 281
7.3.2 Focused tests......Page 285
(a) Bivariate association......Page 287
(b) Spatial association......Page 290
Part D Hypothesis testing and spatial autocorrelation......Page 293
8 Hypothesis testing in the presence of spatial dependence......Page 295
8.1 Spatial autocorrelation and testing the mean of a spatial data set......Page 297
8.2.1 Pearson’s product moment correlation coefficient......Page 300
8.2.2 Chi-square tests for contingency tables......Page 305
Part E Modelling spatial data......Page 309
9 Models for the statistical analysis of spatial data......Page 311
9.1 Descriptive models......Page 314
(a) Models for data from a surface......Page 315
(b) Models for continuous-valued area data......Page 319
(c) Models for discrete-valued area data......Page 326
9.1.3 Models with several scales of spatial variation......Page 328
9.1.4 Hierarchical Bayesian models......Page 329
9.2.1 Models for continuous-valued response variables: normal regression models......Page 334
9.2.2 Models for discrete-valued area data: generalized linear models......Page 338
(a) Adding covariates to hierarchical Bayesian models......Page 342
(b) Modelling spatial context: multi-level models......Page 343
10.1 Models for representing spatial variation......Page 347
(a) Trend surface models with independent errors......Page 348
(b) Semi-variogram and covariance models......Page 349
(c) Trend surface models with spatially correlated errors......Page 353
10.1.2 Models for discrete-valued variables......Page 356
10.2 Some general problems in modelling spatial variation......Page 360
10.3 Hierarchical Bayesian models......Page 361
11.1.1 The ‘classical’ approach......Page 372
11.1.2 The econometric approach......Page 375
(a) A general spatial specification......Page 377
(b) Two models of spatial pricing......Page 378
11.2 Some applications of linear modelling of spatial data......Page 380
11.2.1 Testing for regional income convergence......Page 381
(a) A logistic model with spatial lags on the covariates......Page 383
(b) Autologistic models with covariates......Page 386
11.2.3 Multi-level modelling......Page 387
11.2.4 Bayesian modelling of burglaries in Sheffield......Page 389
11.2.5 Bayesian modelling of children excluded from school......Page 398
11.3 Concluding comments......Page 400
Appendix I Software......Page 401
Appendix II Cambridgeshire lung cancer data (observed and expected counts and ward neighbours)......Page 403
Appendix III Sheffield burglary data......Page 407
Appendix IV Children excluded from school: Sheffield......Page 413
References......Page 416
Index......Page 446