Key concepts 1 GIS - Spatial Data Analysis: An Introduction for GIS Users - page 14

sources and tools may be part of an essential prior step to spatial analysis. In some
contexts, information may be available on road networks and segments of properties
that are located on the roads, but not on the specii c location of properties. In such
cases, given this knowledge, the locations of individual properties can be predicted
through address interpolation (or address matching). h e concept is illustrated in
Figure 2.5, where the coordinates of the road junctions are known and the positions
of properties in the segment are interpolated given that odd-numbered houses are
located on the eastern side of the road while even-numbered houses are located to the
west. h e interpolated position of number 16 is given. Of course, such an approach may
give misleading results where the size of properties, or the spacing between them, varies.
An introduction to geocoding and related issues is provided by Longley et al. (2005a).

Spatial scale

2.7

Spatial data analysis is dependent on the sample size, density, and, where relevant, the
level and type of aggregation (i.e. ways of spatially grouping values; an example is
counting the population within dif erent administrative zones). h e level and type of
aggregation are the subject of Section 4.9. h is section discusses the issue of spatial
scale. In this context spatial scale is dei ned as the scale at which the property of inter-
est varies (but there are many dei nitions and Lloyd (2006) discusses some of these).
For example, in a mountainous landscape, elevation values may dif er a great deal
between one location and another over very short distances. On a river l ood plain
elevation values may, in contrast, dif er very little over even very large distances. In the
former case, the spatial variation may be described as being of a i ne scale or a high
frequency. In the latter case, the spatial variation may be considered coarse scale or

258198, 281685

Road junction
House

27

28

No. 16
258209, 281655

1

2

258229, 281621

Figure 2.5 Geocoding: address interpolation.