Abstract

AbstractA conceptual framework for conversion of data between geographical units is developed. It is used to examine the construction of geographical conversion tables, which have taken many forms and have recently been stimulated in the UK by government emphasis on neighbourhood profiling. Where boundaries of different geographical systems overlap, then conversion of data from one system to another involves approximation. In this case non‐hierarchical geography conversion tables are used and data conversion is equivalent to statistical synthetic estimation. Errors arise both in the construction of geographical conversion tables and in the approximation when converting data between overlapping geographies. A typology of errors when estimating data by geographical conversion is complemented by empirical measurement for a variety of UK examples. Data conversion using weighted sums of source unit data is shown to be more accurate than the allocation of data for whole source units to single target units. Measures of quality for estimates from data conversion are necessary and are proposed. Copyright © 2002 John Wiley & Sons, Ltd.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call