BBC report on womens cycling & fatalities
- Thread starter Crackle
- Start date
Page may contain affiliate links. Please see terms for details.
Origamist
Legendary Member
I've been looking at the BBC report a bit more closely and it is sloppy.
For example:
"Many of the fatalities involving cyclists happen in collisions with a heavy goods vehicle (HGV). This year, seven of the eight people killed by lorries in London have been women.
Considering that women make only 28% of the UK's cycling journeys, this seems extremely high." (my bolds)
What is not mentioned is that women make up a higher percentage of London's cycling journeys (37% in 2006/7). The difference in proportions is crucial if you are examinnig HGV/cyclist collisions in London from a gender perspective.
For example:
"Many of the fatalities involving cyclists happen in collisions with a heavy goods vehicle (HGV). This year, seven of the eight people killed by lorries in London have been women.
Considering that women make only 28% of the UK's cycling journeys, this seems extremely high." (my bolds)
What is not mentioned is that women make up a higher percentage of London's cycling journeys (37% in 2006/7). The difference in proportions is crucial if you are examinnig HGV/cyclist collisions in London from a gender perspective.
MacB
Lover of things that come in 3's
- Location
- Farnborough, Hampshire
Origamist said:
that's exactly the sort of thing I was getting at with my query around stats. One that would be resolved by a national database that listed details of all accidents.
MacB
Lover of things that come in 3's
- Location
- Farnborough, Hampshire
garrilla said:
it matters a lot when doing comparative analysis, you need to be comparing like for like.
MacB said:
Funnily enough, I'm an econometrician/statistician so I understand the role quantatitive analysis.
Firstly, the article is not a technical report. Its a peice of journalism. Its using some basic statistics to thread a report together.
Secondly, in reference to the death in Leeds the author makes the assumption that national figures are likely to be of similar proportions to the London figure.
Thirdly, it offers no descriptive statitics so you could make a relevant comparison even if it were comparing like with like.
If you're going to split hairs over statistics then do it with some vigour...
My personal observation is that many drivers, particularly 'professional' drivers accord women cyclist less respect and room than male cyclists. I do a sheepdog thing with Susie, and it's not because she has less road sense or because she's not fit. I do it because (some, by no means all) truck drivers and taxi drivers look at me (and I look at them) and they think better of taking the piss.
MacB
Lover of things that come in 3's
- Location
- Farnborough, Hampshire
garrilla said:Funnily enough, I'm an econometrician/statistician so I understand the role quantatitive analysis.
Firstly, the article is not a technical report. Its a peice of journalism. Its using some basic statistics to thread a report together.
Secondly, in reference to the death in Leeds the author makes the assumption that national figures are likely to be of similar proportions to the London figure.
Thirdly, it offers no descriptive statitics so you could make a relevant comparison even if it were comparing like with like.
If you're going to split hairs over statistics then do it with some vigour...
it was a criticism of the journalism and, assuming the above is correct, you know full well why it would matter.
Origamist
Legendary Member
garrilla said:Funnily enough, I'm an econometrician/statistician so I understand the role quantatitive analysis.
Firstly, the article is not a technical report. Its a peice of journalism. Its using some basic statistics to thread a report together.
Secondly, in reference to the death in Leeds the author makes the assumption that national figures are likely to be of similar proportions to the London figure.
Thirdly, it offers no descriptive statitics so you could make a relevant comparison even if it were comparing like with like.
If you're going to split hairs over statistics then do it with some vigour...
So you agree with me...Rope-a-dope eh, garilla?!
I used to be more rigorous with this kind of thing on these boards, but got tired of being branded a pedant...
My point, ill made as usual, was that it was impossible for the journo to make a correct statistical appraisal in under 500 words. However, because the significance between the two metrics was huge, it was OK to make this kind of simple statistical conflation - those in the know should see that, those who don't wont care.
There is a major problem with peices like this. SHould the journo write the story without the evidence or, for the sake of getting issues raised, do we allow a bit of technical slippage?
Trust me, if it was an academic paper or government report I would already have fired off a letter. It was a small journalistic item that made a fair point badly, so I'm far less concerned.
There is a major problem with peices like this. SHould the journo write the story without the evidence or, for the sake of getting issues raised, do we allow a bit of technical slippage?
Trust me, if it was an academic paper or government report I would already have fired off a letter. It was a small journalistic item that made a fair point badly, so I'm far less concerned.
garrilla said:
Thanks for the pissing contest. What was that I heard in my ears (**worth their salt...**)
garrilla said:
Why would you want the issue raised at a national level? There is clearly a problem in London of HGVs v Cyclists but elsewhere?
garrilla said:
I think that's a silly attitude to have. The magazine has a big readership. There's already been a campaign around here because of the "jounalistic bad points" that regularly crop up.
Lazy-Commuter
New Member
- Location
- Bottom Right Corner
A point that bothers me somewhat about all this and I'd be interested to hear what the statistical heavyweights have to say ..
Isn't 8 rather a small sample size? Couldn't we just be seeing one of those strange quirky runs? I'd be interested to know what the figures for say last year were (though not interested enough to actually look it up
) .. surely if there (for example) 15 fatalities in 2008 and 14 of them were male, over the two years you'd have 15 males and 8 females out of a total of 23 total fatalities, which would be closer to the ratio of male:female commuters.
I suppose the question I'm asking is: when does a "run" like this become signficant?
Isn't 8 rather a small sample size? Couldn't we just be seeing one of those strange quirky runs? I'd be interested to know what the figures for say last year were (though not interested enough to actually look it up
) .. surely if there (for example) 15 fatalities in 2008 and 14 of them were male, over the two years you'd have 15 males and 8 females out of a total of 23 total fatalities, which would be closer to the ratio of male:female commuters.I suppose the question I'm asking is: when does a "run" like this become signficant?
Similar threads
