Re: Lamps and brakes

"Wally" <atdot@xxxxxxxxxxx> sd / msg
<vVB0f.100078$s83.63984@xxxxxxxxxxxxxxxxxxxxxx> dtd Tue, 04 Oct 2005
20:49:32 GMT:

>inductive reasoning begs its own question - it has no substantive
>foundation beyond the assumptions that are made when it is invoked. And you
>can't cite those same assumptions to validate their invocation!

In populations it is rarely possible to have a sample of 100%.
Statistical sampling is therefore the norm. Less reliable than a 100%
sample, but there you go.

>> Where no conflicting
>> evidence exists, and where it is based on several samples, it is the
>> basis of many of the decisions we make.

>So what? Lack of conflicting evidence doesn't make it true.

Indeed. It does, however, make it substantially more likely than any
other conclusion.

>> I believe Dr Raven is better qualified in this matter than I am.
>Is his being a Doctor-of-something relevant?

Probably. He has more formal training in scientific research than I
do.

>>> I suspect that you have completely failed to grasp my point.

>> Quite possibly. Since you claim your point is not that you dispute
>> the evidence, or that you suggest that your town is different, but
>> that you are - what? an outlier?

>I don't dispute that the evidence you mention exists, but I question the
>validity of asserting that it yields an accurate picture of the relative
>safety of the cycle paths and roads in my town.

So you think your town is as likely to be different from all those
studied as is the same? Interesting. Me, I think if every time they
study a town they get the same result, the result is quite likely to
be typical of all towns. To each his own, I suppose.

>>> So what? Are you going to tell me that 'weight of evidence' makes the
>>> contention (that the cycle paths in my town are more dangerous than
>>> the roads) 'more true'?
>> No, but absence of conflicting evidence makes it more likely.
>What does 'likely' mean?

I don't think I can be much clearer. Every time a study has been
conducted it has found that the roads are safer. No study of which
I'm aware has found otherwise. As far as I can tell the balance of
probabilities is that the same will apply anywhere, and I will be of
that opinion right up to the moment that other evidence becomes
available.

>>>> The reasons for the problem are easily understood.
>>> Now, that's interesting. What are the reasons for the problem?
>> See http://en.wikipedia.org/wiki/Cycle_path for a handy diagram.

>That's not a diagram, it's a photograph. It looks like a road. How far does
>it go? What is at the ends?

The diagram is in the section lower down -
http://en.wikipedia.org/wiki/Cycle_path#The_safety_of_segregated_cycle_facilities

>>> Er, that doesn't answer the question - is M-K a representative
>>> sample?
>> It is *a* sample, one of several. All have the same result.
>The same result to the same degree? The same proportion of accidents on
>roads to cycle paths in every town where there has been a study? Or is there
>variance from one to another?

No, they have different magnitudes but agree as to sign.

>> Are you
>> saying that researchers have chosen only to study unrepresentative
>> towns?

>Are you saying that research in which junctions with roads seem to
>predominate is representative of the safety of cycle paths?

I am saying that wherever the relative safety of cycle paths has been
studied, the result has been that they have a worse record than roads,
with the exception of rural routes, where the distances between
junctions are obviously much higher, as are (typically) traffic
speeds.

>> Am I? Well well. I thought I was stating that, in general, the
>> evidence indicates it's safer on the roads. I wasn't aware I'd been
>> any more specific than that.

>Ah, must have been only Tony that said I would be safer on the roads than
>the cycle paths. So, do you think that the the cycle paths here are more
>dangerous on the roads, but that I'll be safer on the cycle paths? About the
>same? Unknown?

I'd say that the balance of research is such that your starting
premise should be that the roads are likely to be safer. And that you
should be prepared to challenge the natural and widespread assumption
that a cycle path is "obviously" safer than the road.

>http://www.17beechroad.freeserve.co.uk/WarringtonCycleCampaign/facility-of-the-month/index.htm

>Oh, we ain't got nuffink like that. So, they have guys going around,
>painting pictures of little bikes on the roads? Wiv white lines up the side
>of 'em? How is that supposed to segregate bikes and motor vehicles? Haven't
>I already said somewhere in this thread that I'm not aware of this town
>having *any* of that cycle lane stuff?

That is evidence of widespread cluelessness (also a good laugh,
judging from my lad's hysterical reaction to some of them).

>> I use the word many because I do not have enough data on which to base
>> a more confident statement.
>Uh-huh...
>> And because it is alleged that good cycle
>> facilities do exist.
>Really?

Yup. Or so it is said. I am waiting until I see one...

>> A path for cycles separate from the carriageway.

>Then why did you cite a bunch of data in another post which, aside from M-K,
>seemed to concentrate on intersections with roads?

Ah, well, you see, most towns are built around a network of roads. You
typically can't get from anywhere to anywhere else without either
crossing them, riding on them, negotiating a junction, tolerating an
underpass full of chavs, or going over a bridge. In other words, you
either get extra danger or extra work.

>Until that appeared, the
>proposition being propounded was that 'cycle paths are more dangerous than
>roads'. However, your list of references leaves me wondering if it's
>actually the *junctions between roads and cycle paths* that are dangerous.

Indeed. And sufficiently so as to negate the reduced risk between
junctions (especially since the danger between junctions on the road
is typically low anyway).

>Indeed, I find myself wondering if it's the case that cars veer off the
>roads at junctions and collide with cyclists, or whether cyclists fail to
>stop at junctions and place themselves in the paths of motor vehicles.

Neither. Cyclists negotiate the junctions and motorists violate their
right of way. Right of way violations are a common cause of cyclist
injury at junctions (on or off road). Same for pedestrians (rule 146
is widely ignored).

Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken
.

Loading