Hacker News new | ask | show | jobs
by awjr 3348 days ago
It started by looking at http://datashine.org.uk and the sister commute site http://commute.datashine.org.uk/

From which I derived this spreadsheet (ongoing) : https://docs.google.com/spreadsheets/d/1zSLXKO0zRBHeHcunKv5B...

I then downloaded WU03EW MSOA data into this fusion table: https://fusiontables.google.com/DataSource?docid=1sNyIr6EGEE...

From which I've extracted values I am interested in.

However I find it painful to work with.

I now have a city to msoa lookup as well as the population centroids from the datashine commute site (thanks to them for providing them to me)

I am now looking at commuters within a city, those coming from the outside, as well as those travelling across the city.

In particular I'm seeing modal split analysis which indicates 100%+ levels of cycling than being used by the council for transport policy. This may be down to the fact Bath and North East Somerset is a mixed urban/rural county.

I need to find MSOA population densities per km2 (have a source please link!) to be able to determine if there is a correlation between low density and car use. (I suspect it is true given our poor public transport.)

I've now downloaded all the data into a Postgres db to allow me to play with it better and want to paint a national picture.

Once I have this done at MSOA level I want to look at doing this at LSOA level as the accuracy is that much better.