Add new comment

As, perhaps, the person who can lay most claim to being the originator of the current output areas (but not working for ONS) I'd like to comment on a couple of threads emerging from this discussion:

Firstly, I would be quite willing to see us creating an output geography again from scratch, but only in return for some major, unambiguous benefits all round . In that category I would consider a really good match to Mastermap, or perhaps to a set of stable postcodes (neither of which we are sure can be delivered at present), but not just so that we can have a different set of small areas with fundamentally the same characteristics as the ones we already have. That route guarantees no clear benefit to anyone and cannot ultimately resolve any of our likes or dislikes about the present OAs. On the contrary it would prevent geographically consistent comparison over time (a massive lost opportunity) and would mean that users would have to reconstruct all their GIS applications based on the census (costly and inefficient), without any gains of corresponding magnitude.

Secondly, the issue of wards has of course cropped up and they have even been suggested as the geography to hold onto for the sake of consistency. Contemporary wards are the single most important geography for local government. However, they are renamed, recoded and redrawn continually and at different dates in different parts of the country; they range in size from under 1,000 to over 30,000 and there is no freely available, updated, boundary set. Perhaps their greatest failing is that people think they understand them when they don't, leading to massive confusion as to whether ward data are being matched to the correct codes and boundaries. I have no problem with the production of ward statistics as a secondary output product, but they are deficient in almost all important regards as the basis for an official statistical geography.