TfL publish bus accident data

[snoggle]

After a lot of pressure from campaigners and the London Assembly TfL have fulfilled their promise to publish accident data for London Buses. It's for the first 3 months of this calendar year rather than Q1 of the financial year (April to June).

link to data

I've not looked at the data in any great detail but there are 4 sheets giving different "cuts" of the data plus the data itself. It will be interesting to see what follows from the release of this information.

It may be prudent if people did not go into "operator x are useless" or "route y is the death route" type comments or observations. This is three months worth of data which is a small time window and there is no attempt to determine cause or responsibility. Note also that we are talking about the depths of winter plus we had exceptionally wet weather conditions. It is also worth noting the words on the Introduction worksheet and TfL's comments about trying to draw conclusions from the data.

[danorak]

I do worry a bit about this sort of raw data being published without any sort of context: those looking for a scaremongering headline are going to take no notice of that introduction.

[snoggle]

Jun 16, 2014 11:34:22 GMT danorak said:

I do worry a bit about this sort of raw data being published without any sort of context: those looking for a scaremongering headline are going to take no notice of that introduction.

I agree there is a risk but people who do scaremonger can often be challenged by others if everyone's using the same data. Not having the data makes it easier for people to rant on because no one can really challenge them on the facts. At least now some of the facts are published and any stupid comparisons or conclusions will be shown up for what they are.

The problem with "context" is that it's probably different for every single incident so people could drown in the data trying to take "context" into account and getting nowhere very fast. Further some of the context may not be clear, or known or it may involve individuals who don't deserve to have their name dragged into public debate or there may be legal processes underway in some cases. I would be surprised if there is any great trend or theme to the accident data that isn't explained by something obvious like there being lots of people in an area. I doubt there are "killer drivers" or "murder buses" on the rampage or "death bus garages" .

[John tuthill]

Jun 16, 2014 12:19:06 GMT snoggle said:

Jun 16, 2014 11:34:22 GMT danorak said:

I do worry a bit about this sort of raw data being published without any sort of context: those looking for a scaremongering headline are going to take no notice of that introduction.

I agree there is a risk but people who do scaremonger can often be challenged by others if everyone's using the same data. Not having the data makes it easier for people to rant on because no one can really challenge them on the facts. At least now some of the facts are published and any stupid comparisons or conclusions will be shown up for what they are.

The problem with "context" is that it's probably different for every single incident so people could drown in the data trying to take "context" into account and getting nowhere very fast. Further some of the context may not be clear, or known or it may involve individuals who don't deserve to have their name dragged into public debate or there may be legal processes underway in some cases. I would be surprised if there is any great trend or theme to the accident data that isn't explained by something obvious like there being lots of people in an area. I doubt there are "killer drivers" or "murder buses" on the rampage or "death bus garages" .

How about a remake of Stephen King's 'Christine', replace the Plymouth Fury with a Borisbox?