Add new comment

The coment "Should we not invest in algorithms to convert old data to new boundaries, rather than locking ourselves into something that might have been good when conceived but doesn't fit now."
Is an interesting one -THe idea is worth considering, but as with David Martins post, I would ask how the benefits balance out.

Certainly the value of older data is lessened if you consider Censusses. This makes restructuring on the face of it a good option, but the number of datasets that would be restructured, leading to confusion when looking at older data reports, would be significant. In addition, the increasing use of regular and more frequent data means that more frequent reconfiguring could be very costly, and the decision on when it should be done a difficult one.

Consequently the 2001 Census makes sense as a 'review point'. This is something that I hadn't thought sensible until considering this, but the practicality seems to outweigh the advantage.

And what of the advantage - the reason is driven by a geography that 'doesn't fit now'. I'd ask the question 'Why?' if it is changes to population, etc. the polciy would handle that. If it is impractical boundaries, then that is nothing new, they always were problematic (no fault of David's work).

Previously a stumbling block in restructuring was the confidentiality restrictions from overlaid geographies. Is that still a problem?

Finally, the point OAs was flexibility, in order to create different user areas. In theory this would overcome the problem of becoming out-of-date, but it appears not to have done.

The question remains 'Why change?' and the answer remains, it seems, because they were not sufficiently fit for purpose in the first place, rather than things have changed around them.