GIS in Historical Research: The North-West Shropshire Tithe Maps and Apportionments



Capturing the Data Digitally

The Entity-Relationship Diagram

Data Normalisation

The Database Table Structure

Working with the Database


Working with the database

The construction of the database is at an early stage. Hordley Parish data has been entered; using this data, some idea of the usefulness of the database can be gained.

Mapping the distribution of landownership reveals a distinction between the northern (Hordley township) and southern (Bagley township) parts of Hordley Parish. With minor exceptions, Hordley is owned by the Lord of the Manor and farmed by tenant farmers; Bagley is owned by a large number of small landowners. Hordley has only 25% of the houses in the parish; workers live in the farmhouse. Bagley's population live in their own houses, as nuclear families. Hordley has the Anglican church; Bagley has the Primitive Methodist chapel. By inference Bagley has social traits of individualism associated with nonconformism and the family household. Hordley represents the 'old' style socio-economic structure; Bagley represents the 'new' style economy and society with ownership spread at least to the middling classes.

Although Hordley township had a landlord and tenant economy, it should not be inferred that tenants are necessarily exploited. As the maps below show, several tenants were able to accumulate sufficient funds to buy land in Bagley township. The landlord, for whatever reason, did not expropriate the entire surplus generated by the tenants. This suggests a symbiotic landlord-tenant relationship, perhaps akin to the New Zealand sharemilker situation, rather than a Marxist/Structuralist dialectic model.

The relationship between tithe per acre as a measure of production and parcel size is analysed by calculating correlation coefficients, firstly for all tithe areas and secondly for subsets of different land uses (arable, meadow/meadow and pasture, pasture and houses). For all parcels r2 =-.0707; this is not significant at the .05 level, indicating no correlation between parcel size and rent per acre. Splitting the data by land use, r2 for arable is -.02024, for meadow/meadow and pasture -.3111, for pasture-.1888 and for houses .244788. None are significant at the .05 level. Such weak correlations as these indicate very slight diseconomies of scale for uses other than housing; as field size increases productivity decreases. It is difficult to suggest reasons for this other than larger fields losing the weather protection of hedges, resulting in more wind throw in corn crops and more exposure for animals. For houses there is a weak positive correlation; rent increases with size of parcel. Table 1 shows the rent for parcels in ascending order. The descriptions of the use of the parcels in the landtype column suggest that the increase is likely to be due to the quantity of buildings on the parcel, as much as their quality.

Table 1. Rent payable and use of land with houses.



House,Garden and Malthouse


House, Garden and Buildings


House, Buildings and Garden


Smiths Shop, House etc.


House and Garden


House and Gardens


House, Garden and Road


House, Garden, Buildings, Fold


House, Garden and Buildings


House and Garden


House, Garden, Buildings, etc


House,Buildings etc.


House, Buildings and Part of Garde


House,Buildings,Fold etc.


House,Buildings and Garden


House,Garden,Fold and Buildings,et


House,Buildings, etc


House,Garden,Fold and Buildings,et




Variation in tithe per acre may be due to factors other than parcel size, e.g. slope, aspect, soil type, occupier and owner. The Hordley data is too small a sample to test the whole range of factors. When the larger data set is available a multilevel statistical analysis will be undertaken using the all the factors.