The Difficulty of Linking Two Differently Aggregated Spatial Datasets: Using a Look-up Table to Link Postal Sectors and 1991 Census Enumeration Districts

The use of postal geography as areal units has developed significantly since its first introduction in the 1971 Census of Scotland. The 1987 Chorley report advocated the use of postal codes as the standard areal unit for publication of geographic data across the board. The change to a postal base in the census of England and Wales finally took place in 2001. Aggregation of population data is essential, both to protect the identity of individuals and make the data manageable. The question however remains, whether the independent aggregation of two similar datasets covering the same geographical area makes the data two separate information sources, or whether they can be successfully be linked together and used as one. Through the comparison of two aggregated areal datasets of British population based statistics, this paper examines the reliability of commercially produced, undocumented data, with the use of a look-up table linking postal sectors to enumeration districts of the 1991 Census of British population. The investigation finds that the ability to link the Experian dataset is to the census is questionable. The two datasets contain many obvious and significant differences when linked, it can be concluded that look-up tables are a poor and inaccurate way of linking differently aggregated spatial datasets.