Bath & NE Somerset Council

Revealing patterns in Bath residents’ attitudes to a new casino

Geofutures were commissioned by Bath & North-East Somerset (B&NES) council’s Economic Development Service to analyse the results of a Voicebox survey canvassing opinion among B&NES residents about the proposed casino development in Bath.

The survey was carried out by market survey consultancy Snap prior to the February 2007 announcement that B&NES were among the local authorities which had successfully won a recommendation from the Casino Advisory Panel as a location for a new casino. The analysis by Geofutures compared its results with national datasets, and set them in the context of geo-demographic differences between respondents. Results were then visualised and animated in a mini-application to make them easy to understand and base decisions upon.

For any type of quantitative analysis, applying results gathered at a large spatial scale (like national opinion surveys) to local contexts creates issues of comparison. In some cases, the rounding-off of figures derived from large geographic zones (e.g. a region or county) makes it statistically unsafe to report findings at finer scales (e.g. a district or ward). Even where this is not being attempted, opinions vary between different locations at a scale which national surveys cannot convey.

It’s vital, therefore, to compare local opinions with national surveys in statistically appropriate ways. This ensures that local results can be understood in relation to broader trends, highlighting areas of difference from national figures in order to inform policy.

The science of geo-demographics offers one means of making appropriate national and local comparisons and drawing conclusions from the results. In this study Geofutures used the geo-demographic system Output Area Classification (OAC), which applies neighbourhood classifications to national census data. These groupings can be sub-divided into increasing levels of detail, but for the purposes of this study the seven key classifications were used to analyse results.

The spatial distribution of the B&NES survey respondents and non respondents revealed variations in responses between different locations, offering a valuable benchmark for future monitoring of social impacts and attitudes.