🚀 go-pugleaf

nmm1@cam.ac.uk wrote:
> 
> The actual data is indirected to yet again.  If they are in Ashton's PhD
> thesis, they should be findable, but the first paper may well no longer
> exist.  Anyway, there is nothing on the sample size or methodology, and
> my guess is that the former is small, given that it comes "mainly from
> data obtained in at-the-scene pedestrian accident studies conducted at
> the Accident Research Unit, University of Birmingham".

  It used to be available, some of the sample sizes were admitted to be 
very small, and unrepresentative, but suitable for the original purpose 
(how to improve car bonnet design 30 years ago).

In article <h9gfos$tfp$1@news.eternal-september.org>,
Nick Finnigan  <nix@genie.co.uk> wrote:
>> 
>> The actual data is indirected to yet again.  If they are in Ashton's PhD
>> thesis, they should be findable, but the first paper may well no longer
>> exist.  Anyway, there is nothing on the sample size or methodology, and
>> my guess is that the former is small, given that it comes "mainly from
>> data obtained in at-the-scene pedestrian accident studies conducted at
>> the Accident Research Unit, University of Birmingham".
>
>  It used to be available, some of the sample sizes were admitted to be 
>very small, and unrepresentative, but suitable for the original purpose 
>(how to improve car bonnet design 30 years ago).

Thanks very much.  Yes, it was pretty clear that the paper was
attempting to describe the major factors and give an indication
of their magnitude.  As such, there was nothing wrong with it.
There's nothing wrong with those figures, as rough indications of
the sort of effect, either.

My objections are solely to the way that it has been abused, to the
way that a completely bogus conclusion is touted as authoritative,
and to the consequent damage to road planning.


Regards,
Nick Maclaren.

nmm1@cam.ac.uk wrote:
> There are some figures commonly bandied around by the DfT, RoSPA and
> other organisations and people who ought to know better, about the
> proportion of pedestrians who are killed when hit by cars at various
> speeds.  I have been looking for it for some time, and it seems to have
> been scanned in and put on the Web:
> 
>     http://www.dft.gov.uk/foi/responses/2005/nov/203040message/paperaboutthedepartments20302445

  What you want is
http://www.dft.gov.uk/foi/responses/2005/nov/203040message
which has the original written response to someone who obviously sent a lot 
of emails to the department around 2005 (and probably wore some dodgy ties)

As well as the above paper, there is an Ashton 1980 paper which gives more 
details of his thesis.

On Thu, 24 Sep 2009 20:54:00 +0100, Nick Finnigan put finger to
keyboard and typed:

>nmm1@cam.ac.uk wrote:
>> There are some figures commonly bandied around by the DfT, RoSPA and
>> other organisations and people who ought to know better, about the
>> proportion of pedestrians who are killed when hit by cars at various
>> speeds.  I have been looking for it for some time, and it seems to have
>> been scanned in and put on the Web:
>> 
>>     http://www.dft.gov.uk/foi/responses/2005/nov/203040message/paperaboutthedepartments20302445
>
>  What you want is
>http://www.dft.gov.uk/foi/responses/2005/nov/203040message

That's a 404 for me.

Mark
-- 
Blog: http://mark.goodge.co.uk
Stuff: http://www.good-stuff.co.uk

In article <h9gior$nlf$1@news.eternal-september.org>,
Nick Finnigan  <nix@genie.co.uk> wrote:
>
>> There are some figures commonly bandied around by the DfT, RoSPA and
>> other organisations and people who ought to know better, about the
>> proportion of pedestrians who are killed when hit by cars at various
>> speeds.  I have been looking for it for some time, and it seems to have
>> been scanned in and put on the Web:
>> 
>>     http://www.dft.gov.uk/foi/responses/2005/nov/203040message/paperaboutthedepartments20302445
>
>  What you want is
>http://www.dft.gov.uk/foi/responses/2005/nov/203040message
>which has the original written response to someone who obviously sent a lot 
>of emails to the department around 2005 (and probably wore some dodgy ties)
>
>As well as the above paper, there is an Ashton 1980 paper which gives more 
>details of his thesis.

Indeed.  Thanks very much.  The page 404s, due to the DfT's damn-fool
Web serving, but I can get there by adding a trailing '/'.

The other paper gives the data, but the methodology is still in the
possibly unobtainable 1977 paper, as I assume is any more detail
on the method of selection.

But it DOES confirm my guess that those graphs are distribution fits,
and so the conclusions draft by the DfT, RoSPA etc. are bullshit,
and arguably propaganda as well.

Please pass my thanks for extracting this information out of what I
assume was a very reluctant DfT to the person with dodgy ties :-)


Regards,
Nick Maclaren.

nmm1@cam.ac.uk wrote:
> 
> Please pass my thanks for extracting this information out of what I
> assume was a very reluctant DfT to the person with dodgy ties :-)

  The person with the dodgy ties is no longer with us, but I expect the DfT 
were very reluctant, even if they didn't understand how grossly they were 
misinterpreting a single smoothed graph with inadequate labels.

nmm1@cam.ac.uk wrote:

> Anyway, there is nothing on the sample size or methodology, and
> my guess is that the former is small, given that it comes "mainly from
> data obtained in at-the-scene pedestrian accident studies conducted at
> the Accident Research Unit, University of Birmingham".
> 
You might find relevant papers in IRCOBI proceedings; as well as Ashton, 
others to consider are Pedder, Mackay & Gloyns who also worked in the 
Birminhgham Accident Research Unit & used data from Birmingham Hospital over 
much the same period. (Pedder I think focused on motorcycle accidents.)