Dragons' Den backs indicators: another dodgy decision from the dragons on a cycling product

Rich_cb wrote:

Just read the research. Stop pontificating about how you think they did it and just read it.

nowhere does it say that it was "found to be a contributing factor" - the report clearly says "attributed" in the 10% figure (p.34)  - elsewhere it mentions both dark clothing and no lights together (p.23) - and in the conclusion it points to this being a factor on rural roads (p.45) and recommends promotion of both the use of lights and clothing...

so, unlike the report itself, you've taken one variable out of context and turned it into a definitively causal factor

Rich_cb wrote:

beezus fufoon wrote:

shay cycles wrote:

Rich_cb wrote:

You could always read the research. They explain quite clearly where they are getting their data. It's mainly based on the STATS19 data collected by the police, also hospital data, coroners reports and 'on the site' reports.

Well in reality research based mainly on STATS19 needs to be handles with care. The government's own guidance says;

"The factors are largely subjective, reflecting the opinion of the reporting police officer, and are not necessarily the result of extensive investigation. Some factors are less likely to be recorded since evidence may not be available after the event.

While this information is valuable in helping to identify ways of improving safety, care should be taken in its interpretation."

So the data quoted is far from reliable.

So you cannot reasonably claim that dark clothing was a contributary factor in 10% of fatalities - whether a University study says so or not.

 

equally, if it is just raw data, then it is clearly 10% of fatalities who are wearing dark clothing - if it's a different percentage and discouted in some instances, then it's a question of interpreting that data - in which case they will have certain criteria which can be validated and replicated...

I suspect they have just used a certain phrasing - "it is a contributory factor" - to make their common sense assumptions appear scientific...

I wonder whether dark coloured cars are involved in proportionately more collisions - or if they discovered that neon green cars were, would they then count colour as a contributory factor?

I also suspect that we might find that the amount of people who wear dark clothes is upwards of 30% and that they are under represented by the 10% figure - it is of course possible that those who wear hi-viz tabards are more safety conscious and more cautious riders, so the clothing is not actually causal, simply a reflection of other factors - there are so many possibilities and other factors to consider here.

Just read the research.

Stop pontificating about how you think they did it and just read it.

You are the one citing it in support, why don't you read it and explain it? It shouldn't be that hard to give the gist of the argument, i.e. how they arrived at the conclusion and how they allowed for other possible explanations of the data.

I went to download it and decided I didn't feel like giving them all my personal details to be able to see it...I'm assuming they then do let anyone download it rather than then insisting you have to have some professional justification for doing so?

Maybe I'll give it a go at some point. But in the meantime, you are the one citing it in support so seems to me the onus is on you to explain what it says, given they don't just put it up for anyone to read.

It might be convincing and sound, but in the absence of a clear outline of the paper, I don't feel inclined to take the TRL's conclusions on faith. It is, after all, now a private think-tank for-hire, owned by 'members of the transport industry'.

Seems to me that set up might tend to produce reports saying what those paying for the reports want to hear. So the methodology is pretty important.

Rich_cb wrote:

Just read the research.

Stop pontificating about how you think they did it and just read it.

But didn't you comment/elaborate about how you thought they meant it?

Rich_cb wrote:

I don't know exactly what they meant by 'dark clothing' but I'd imagine they meant dark colours and non reflective.

FluffyKittenofTindalos wrote:

...a private think-tank for-hire, owned by 'members of the transport industry'.

Seems to me that set up might tend to produce reports saying what those paying for the reports want to hear...

more likely the sort who need a report to tell them that it's a bad idea to ride an unlit bicycle around country lanes in the middle of the night

ClubSmed wrote:

Rich_cb wrote:

Just read the research.

Stop pontificating about how you think they did it and just read it.

But didn't you comment/elaborate about how you thought they meant it?

Rich_cb wrote:

I don't know exactly what they meant by 'dark clothing' but I'd imagine they meant dark colours and non reflective.

A bit different, they didn't define 'dark clothing' in the report but they did clearly describe their method.

FluffyKittenofTindalos wrote:

You are the one citing it in support, why don't you read it and explain it? It shouldn't be that hard to give the gist of the argument, i.e. how they arrived at the conclusion and how they allowed for other possible explanations of the data.

I went to download it and decided I didn't feel like giving them all my personal details to be able to see it...I'm assuming they then do let anyone download it rather than then insisting you have to have some professional justification for doing so?

Maybe I'll give it a go at some point. But in the meantime, you are the one citing it in support so seems to me the onus is on you to explain what it says, given they don't just put it up for anyone to read.

It might be convincing and sound, but in the absence of a clear outline of the paper, I don't feel inclined to take the TRL's conclusions on faith. It is, after all, now a private think-tank for-hire, owned by 'members of the transport industry'.

Seems to me that set up might tend to produce reports saying what those paying for the reports want to hear. So the methodology is pretty important.

I've provided you with a link to a free full text copy of the report.

I'm not going to read it to you word for word because you can't figure out how to download it anonymously.

I've provided the evidence, the onus is on you to read it before drawing conclusions.

beezus fufoon wrote:

Rich_cb wrote:

Just read the research. Stop pontificating about how you think they did it and just read it.

nowhere does it say that it was "found to be a contributing factor" - the report clearly says "attributed" in the 10% figure (p.34)  - elsewhere it mentions both dark clothing and no lights together (p.23) - and in the conclusion it points to this being a factor on rural roads (p.45) and recommends promotion of both the use of lights and clothing...

so, unlike the report itself, you've taken one variable out of context and turned it into a definitively causal factor

 

Not at all.

Table 7 on page 23 clearly show dark clothes at night to be a contributing factor in 10% of fatal collisions.

It is a separate contributory factor to lights.

The word attributed is used to describe whether the contributory factor was on the part of the cyclist or motorist.

davel wrote:

Weak, subjective and incomplete research is worse than none at all if it's used to justify weird and wonderful leaps in the dark.

I really don't think this can be described as weak or incomplete. It covers a huge numbers of fatal collisions and has examined most if not all of the available data.

Data of this type will always be subjective, there is no way to produce objective data about these sorts of collision.

Finally it's not a leap in the dark to suggest that removing a reported contributing factor could reduce the incidence of fatal collisions.

It's a logical response.

Rich_cb wrote:

davel wrote:

Weak, subjective and incomplete research is worse than none at all if it's used to justify weird and wonderful leaps in the dark.

I really don't think this can be described as weak or incomplete. It covers a huge numbers of fatal collisions and has examined most if not all of the available data.

Data of this type will always be subjective, there is no way to produce objective data about these sorts of collision.

Finally it's not a leap in the dark to suggest that removing a reported contributing factor could reduce the incidence of fatal collisions.

It's a logical response.

I might be going round in circles here, but my point all along has been how logical that response is.

Am I right in that the problem here is the contributing factor - dark clothing?

If so, a lot hinges on the term 'dark clothing', how they've arrived at that definition and what the basis for including deaths under that label is. But then they don't explain how they've done that? (This is where I think ClubSmeds was going)

So how do you fix a problem as defined by a bunch of other people if they haven't told you what the label that they've given the problem actually means?

If we don't know what it means, how do we remove 'dark clothing' here? (I think this is where Beezus was going. ) Do they mean just unlit clothing? Do they merely mean that the cyclist was in the dark? Are they just saying that it was dark or is the problem with the actual clothing? Types of garment? Clothing brands? Does hair colour make a difference? Do peds suffer the same casualty rates? You see what large effects a subtle difference in the label could have?

Next, are reflectives demonstrably the antidote, as opposed to 'light clothing' or just nakedness or a Ronald Macdonald outfit or a portable, head-mounted spotlight? How effective are they? Should cyclists be advised to avoid cycling in the dark if their clothes make drivers hit them? What might the unintended consequences be of applying the suggested fix? Is there a wider problem with drivers not driving to the conditions on unlit roads or elsewhere (I think this is where ktache was regarding shit driving and you were regarding factors beyond your control)?

(I'm not actually being obtuse here - I spent a long time working in performance improvement and the same mistakes regarding 'fixing problems' happen repeatedly, and one of the most common is a lack of understanding and agreement as to what the problem is in the first place. And complex problems are never fixed via 'common sense'. I hear your Swiss Cheese model, but you need some confidence that the slices address the factors, and I lack confidence that they've been defined correctly here.

TL;DR:
1. Can we demonstrate what' dark clothing' means?
2. Can we be confident that those figures were categorised accurately?
3. Can we demonstrate that reflectives address the issue of 'dark clothing'?

Isn't this' no' on all counts?

davel wrote:

TL;DR:
1. Can we demonstrate what' dark clothing' means?
2. Can we be confident that those figures were categorised accurately?
3. Can we demonstrate that reflectives address the issue of 'dark clothing'?

Isn't this' no' on all counts?

1. We can't know definitely what they meant by dark clothing.

It would be reasonable to say that high-vis clothing would be very unlikely to be classified as dark.

In certain lights reflective clothing can appear as dark so there may be a few cases where reflective clothing has been recorded as dark.

Realistically though I would expect almost all people to be able to identity dark clothing correctly.

Overall I think it is perfectly reasonable therefore to assume that dark clothing means just that in the vast majority of cases they've recorded.

2. I have quite a lot of confidence in these figures, they seem to have been very methodically gathered and seem to tally with real world experience.

The fact that they report a high incidence of death at junctions tallies with what I read about on an all too regular basis.

The fact that they found that car drivers were inattentive and drove at excessive speeds mirrors exactly what I experience on an almost daily basis.

3. There is no definite proof of this, it therefore becomes a personal choice. In low light conditions reflective gear is the brightest stuff you can wear, if dark clothing is a problem then it seems sensible to wear the brightest clothing you can.

You're never going to get scientific level proof in these sort of studies, I think this research is about as good as you can get given the data available.

Rich_cb wrote:

shay cycles wrote:

Rich_cb wrote:

You could always read the research. They explain quite clearly where they are getting their data. It's mainly based on the STATS19 data collected by the police, also hospital data, coroners reports and 'on the site' reports.

Well in reality research based mainly on STATS19 needs to be handled with care. The government's own guidance says;

"The factors are largely subjective, reflecting the opinion of the reporting police officer, and are not necessarily the result of extensive investigation. Some factors are less likely to be recorded since evidence may not be available after the event.

While this information is valuable in helping to identify ways of improving safety, care should be taken in its interpretation."

So the data quoted is far from reliable.

So you cannot reasonably claim that dark clothing was a contributary factor in 10% of fatalities - whether a University study says so or not.

 

Have you got any better data? Links to any better research?

No I have no better data, there is unlikely to be any better data. But data based on subjective opinions is unreliable and therefore all of the conclusions drawn from it are unreliable. The statement that the report uses STATS19 data effectively invalidates it. This really is pretty basic scientific or statistical method.

In the absence of "better data" you effectively have no data and no usable data yet insist on defending your unsubstatiated position regarding this report.

shay cycles wrote:

No I have no better data, there is unlikely to be any better data. But data based on subjective opinions is unreliable and therefore all of the conclusions drawn from it are unreliable. The statement that the report uses STATS19 data effectively invalidates it. This really is pretty basic scientific or statistical method.

In the absence of "better data" you effectively have no data and no usable data yet insist on defending your unsubstatiated position regarding this report.

If you accept this is the best data available but reject it then you have to accept that there is no reliable data proving drivers are at fault in the vast majority of fatal collisions.

That is what this data shows, everyone on this thread is desperate to discount the data, I doubt they're so keen to stop blaming drivers for fatal collisions.

Your call.

ktache wrote:

And the investigating officer, who has driven there, who has probably not ridden a bicycle since they were a child, if that (excellent WMP excepted) and who has only heard the justification of the motorist....

Which is why the motorist was found to have been the contributing factor in the vast majority of fatal collisions?

The tribalism on here is pathetic.

beezus fufoon wrote:

The word attributed is used to describe whether the contributory factor was on the part of the cyclist or motorist.

1. table 7-4 is on page 34

2. it is treated as separate on that table, but this is qualified in the text

3. it also clearly explains that "attributed" refers to the attending police officers' judgments

Rich_cb wrote:

Data of this type will always be subjective, there is no way to produce objective data about these sorts of collision.

This is the exact point we've all been making to you from the beginning

[/quote]

So reject this research and all of its findings. That means you can no longer blame drivers for fatal collisions. After all in the vast majority of cases the evidence is subjective.

Who else is going to attributed contributing factors?

It's the best data set available, it backs up the arguments that cyclists have been making for years, if you reject it you have to reject all of it, it's all just as subjective.

Rich_cb wrote:

beezus fufoon wrote:

Quote:

The word attributed is used to describe whether the contributory factor was on the part of the cyclist or motorist.

 

1. table 7-4 is on page 34

2. it is treated as separate on that table, but this is qualified in the text

3. it also clearly explains that "attributed" refers to the attending police officers' judgments

 

Rich_cb wrote:

Data of this type will always be subjective, there is no way to produce objective data about these sorts of collision.

This is the exact point we've all been making to you from the beginning

So reject this research and all of its findings. That means you can no longer blame drivers for fatal collisions. After all in the vast majority of cases the evidence is subjective.

Who else is going to attributed contributing factors?

It's the best data set available, it backs up the arguments that cyclists have been making for years, if you reject it you have to reject all of it, it's all just as subjective.

Firstly, I am neither accepting nor rejecting this data - simply correcting your phrasing when you portray this statistical data as scientific fact.

Secondly, I think it is also a bit simplistic to blame drivers for fatal collisions - they are not the ones manufacturing overpowered fuel-hungry vehicles or facilitating their use in inappropriate road conditions, nor are they responsible for the upkeep of the roads, provision of infrastructure, or the enforcement and setencing of offenders.

Thirdly, this data clearly reflects the perceptions and biases of the police and society in general, particularly in placing the onus on the individual in circumstances where it is very obvious that the wider conditions play a far bigger part than the actions of any single participant. The report itself makes that very clear.

Rich_cb wrote:

ktache wrote:

And the investigating officer, who has driven there, who has probably not ridden a bicycle since they were a child, if that (excellent WMP excepted) and who has only heard the justification of the motorist....

Which is why the motorist was found to have been the contributing factor in the vast majority of fatal collisions?

The tribalism on here is pathetic.

And yet, given that 2 of us die each week through the fault of people in more comfortable and safe, and very polluting, metal boxes, and the transport secretary doors a cyclist and then attempts to blame the cyclist, that tribalism is probably justifiable, indeed necessary.

davel wrote:

And yet, given that 2 of us die each week through the fault of people in more comfortable and safe, and very polluting, metal boxes, and the transport secretary doors a cyclist and then attempts to blame the cyclist, that tribalism is probably justifiable, indeed necessary.

How is tribalism working out for us so far?

Research like this can be used to argue for safer junctions, improvement in HGV cabs, better segregated routes to schools and many, many other things that most cyclists care deeply about.

If we needlessly attack this sort of research we are doing the work of the car lobby for them.

Rich_cb wrote:

davel wrote:

And yet, given that 2 of us die each week through the fault of people in more comfortable and safe, and very polluting, metal boxes, and the transport secretary doors a cyclist and then attempts to blame the cyclist, that tribalism is probably justifiable, indeed necessary.

How is tribalism working out for us so far?

Research like this can be used to argue for safer junctions, improvement in HGV cabs, better segregated routes to schools and many, many other things that most cyclists care deeply about.

If we needlessly attack this sort of research we are doing the work of the car lobby for them.

Show me one post that 'needlessly attacks' the research. I see repeated questioning of its methodology and the logic of the conclusions you're drawing, but no needless attacks.

As for tribalism, what sort of person argues for safer junctions and many, many other things? Well, as someone who does that locally, and knows others who do the same, I know it's the type of single-minded, thick-skinned person who is mightily pissed off with society's car worship and acceptance of cyclists' deaths and who posts on threads like this one.

The Netherlands didn't get where they are with infrastructure by more people strapping on a reflective jacket, another light or a plastic lid. Stop de kindermoord! was just another tribal movement to begin with. How's that worked out?

davel wrote:

Show me one post that 'needlessly attacks' the research. I see repeated questioning of its methodology and the logic of the conclusions you're drawing, but no needless attacks.

As for tribalism, what sort of person argues for safer junctions and many, many other things? Well, as someone who does that locally, and knows others who do the same, I know it's the type of single-minded, thick-skinned person who is mightily pissed off with society's car worship and acceptance of cyclists' deaths and who posts on threads like this one.

The Netherlands didn't get where they are with infrastructure by more people strapping on a reflective jacket, another light or a plastic lid. Stop de kindermoord! was just another tribal movement to begin with. How's that worked out?

Attacking the research for being subjective is pointless and needless.

There is no way of doing this type of research objectively.

It relies on value judgements.

ClubSmed wrote:

I too have worked most of my life in process improvement which is why I too want change but am not willing to accept the "facts" without questioning them or the conclusions if made by leaps of faith.
So often I have encountered local improvements made be a department that has improved their money/time/quality position. On further investigation though the full end to end process has suffered greater though. Plugging a small hole at one point can cause a bigger hole to appear elsewhere unless causality is fully understood.
This is what has been suggested here. What if the excess use of reflective does make cyclists safer at night but because they are not as visible during the day they become less safe. Further more because, as the results of this study show, more cyclists ride during the day the fatalities increase as a result?
I'm not saying that visibility isn't an issue, I question whether the fix to that issue will cause other larger issues or not.

There is absolutely no way of knowing whether there will be knock on effects or not.

So we can either taken no action and avoid any unforseen circumstances or take action based on the best evidence we have available.

I think the latter is a more sensible approach.

Rich_cb wrote:

ClubSmed wrote:

I too have worked most of my life in process improvement which is why I too want change but am not willing to accept the "facts" without questioning them or the conclusions if made by leaps of faith.
So often I have encountered local improvements made be a department that has improved their money/time/quality position. On further investigation though the full end to end process has suffered greater though. Plugging a small hole at one point can cause a bigger hole to appear elsewhere unless causality is fully understood.
This is what has been suggested here. What if the excess use of reflective does make cyclists safer at night but because they are not as visible during the day they become less safe. Further more because, as the results of this study show, more cyclists ride during the day the fatalities increase as a result?
I'm not saying that visibility isn't an issue, I question whether the fix to that issue will cause other larger issues or not.

There is absolutely no way of knowing whether there will be knock on effects or not.

What??! I don't follow this. Did you mean something else? Do you need a sit down or to improve your googling? Other research on different topics exists, as does impact analysis as a discipline. I don't want to appear condescending but as it reads, that statement is just false.

There is other research addressing the stuff that ClubSmeds has suggested as one area to look at. Even if there wasn't, it's possible for it to be commissioned. There's research that suggests that making cycling less convenient or appear more dangerous (use of safety equipment) has the effect of reducing numbers of people who cycle. That sort of thing.

But, in order to limit unintended consequences to begin with, the argument needs to be tied tightly to the research and the research itself needs to be tight, which is the cause for alarm here with the 'dark clothing is a factor in 10% of fatalities therefore reflectives' line of reasoning. Picking apart every word in that is second nature to the nerds who do this sort of thing for a living or, even weirder, a hobby.

Rich_cb wrote:

ClubSmed wrote:

I too have worked most of my life in process improvement which is why I too want change but am not willing to accept the "facts" without questioning them or the conclusions if made by leaps of faith.
So often I have encountered local improvements made be a department that has improved their money/time/quality position. On further investigation though the full end to end process has suffered greater though. Plugging a small hole at one point can cause a bigger hole to appear elsewhere unless causality is fully understood.
This is what has been suggested here. What if the excess use of reflective does make cyclists safer at night but because they are not as visible during the day they become less safe. Further more because, as the results of this study show, more cyclists ride during the day the fatalities increase as a result?
I'm not saying that visibility isn't an issue, I question whether the fix to that issue will cause other larger issues or not.

There is absolutely no way of knowing whether there will be knock on effects or not.

So we can either taken no action and avoid any unforseen circumstances or take action based on the best evidence we have available.

I think the latter is a more sensible approach.

Of course there is with more research and better understanding.
To take your Swiss cheese analogy from earlier, you have one slice with holes in it and you think that by moving it slightly you can stop the lining up of a small hole with the next slice. What you're not doing is looking at the whole thing to make sure that by adjusting to remove that how your not making a larger hole line up.
I already provided a link to a research paper that showed the color of clothing at night to not be a factor, only reflectives. You then stated to interoperate"dark clothing" in a way that there was no evidence to support. That is dangerous assumption and could lead to serious Swiss cheese misalignment. We need to fully understand the current situation, the causes and their effects to the whole to ensure we are not exposed to a hole!

davel wrote:

There is other research addressing the stuff that ClubSmeds has suggested as one area to look at. Even if there wasn't, it's possible for it to be commissioned. There's research that suggests that making cycling less convenient or appear more dangerous (use of safety equipment) has the effect of reducing numbers of people who cycle. That sort of thing.

Exactly! The studies into"shared space" set ups have shown that is you make vulnerable uses more common and less predictable then they become safer as drivers are looking out for them better and therefore more aware.
I'm not saying that we need to go around at night dressed like ninjas, I'm also not sure that the Christmas tree stance is any more helpful though