Welcome to Netrider ... Connecting Riders!

Interested in talking motorbikes with a terrific community of riders?
Signup (it's quick and free) to join the discussions and access the full suite of tools and information that Netrider has to offer.

The "evidence" Vicroads is using to justify Hi Viz

Discussion in 'Research, Studies, and Data' started by robsalvv, Jul 9, 2014.

  1. #1 robsalvv, Jul 9, 2014
    Last edited: Jul 9, 2014
    As some of you would know, mandatory hi viz for learner riders is coming.

    The "evidence" used to support hi viz is as follows:

    The role of conspicuity measures in improving detection of motorcyclists in traffic was examined by Wells et al (BMJ 2004; 328), who reported a 37% reduction in risk for riders who wore reflective or fluorescent clothing.


    In the recommendation arising from the Inquiry into Motorcycle Safety, the Victorian Parliamentary Road Safety Committee paid a good deal of attention to the work of De Craen et al (2011), published by SWOV. De Craen highlights that brightly colored high-visibility clothing is more effective in an urban environments which, in Victoria, is where 80 per cent of 2-vehicle novice rider crashes occur (Victoria Police crash data, 2009-2013).


    Research regarding the role of conspicuity can also be found in Motorcycle Conspicuity – What Factors Have the Greatest Impact? http://www.intrans.iastate.edu/rese...h-reports/motorcycle_conspicuity_ii_w_cvr.pdf. This study found a significant increase in detection distance of brightly cloroured riders in both rural and urban settings. This more recent work may not have been available to the Parliamentary Committee when it made its deliberations as the report was not published until June 2012.

    This thread is to explore the pros and cons of the notion of hi viz and the validity of the research.

    Interestingly both of the first two research papers make mentions of the possible confounds in their population sample which may render their conclusions moot. (Note, Vicroads doesn't mention this!)

    The first paper tries robustly to rationalise the confound away, but it's been roundly condemned by many since it's release. Perhaps we can explore what these confounds mean in the posts ahead. @smileedude@smileedude and other research heads, I'm looking at you! It really is an important notion to understand and why statistical studies on this topic are generally flawed and why a proper randomised trial is the only way to answer whether hi viz has any genuine efficacy.

    The third research paper is a simulator study asking participants to press a button when they first noted a bike. Yes, conspicuous bikes were noted further away than non conspicuous bikes, but the interesting fact that appears to have gone unnoticed about this study is that when given the task of noting bikes, ALL bikes were noted, even the dark ones. To me it reinforces the well understood human condition of you see what you expect to see. Given that most drivers don't expect to see bikes, they DON'T even when in plain sight. This is a common experience amongst riders and why we advise each other to "ride around like we're invisible", because for a lot of the time, we are invisible through NO PHYSICAL APPEARANCE FAULT OF OUR OWN.

    Clearly Vicorads has made the leap that if a bike is seen then it will be safer. However, the simulator study doesn't say any such thing.

    The other thing is even if we are seen, other cognitive issues, the most predominant one known as "time to arrival illusion" (it's even mentioned on the TAC Spokes site), means that drivers often incorrectly assess the bike's speed and distance, so they drive out into our path assuming we are further away - this is the classic "SMIDSY" type incident often leading drivers to believe we were speeding. Hi Viz will not help this cognitive issue in any way.

    What will help address this cognitive issue is awareness and education. For example, when I see a bike in the distance (i.e. awareness, I'm expecting and looking for bikes), I don't judge it's distance by how far away it looks - I scan the road all the way up to the bike and get a much truer sense of how far away it is. This is NOT common practice, but it should be (ie education).

    I will say this though, despite the fears of thin edge of the wedge stuff, I don't believe this is a first step move towards mandating all riders in hi viz. I think the notion comes from a genuinely well meaning place to help give our most vulnerable road users, novice riders, every edge they can on the road. As we know, on 3AW recently, the announcer expressed the opinion of many, even if hi viz doesn't help, it can't hurt. This is a very hard notion to shift but it is not a good notion to base new regulation on.

    Ok, over to you guys.

    • Like Like x 3
  2. Firstly I hate the idea of mandating any personal safety gear. I don't think we should be forced to wear helmets or seat belt, despite the clear evidence of the effectiveness of these devices, although the government is more than welcome to tell us the benefits of such things. Like wise with high vis, I am opposed to it, not because it might not help but because the government has no business sticking its nose in to riders apparel. Though it could be argued that high vis is not a personal safety device as it is mitigating against multi-vehicle accidents.

    However I do think it would probably be moderately effective and potentially reduce the new rider toll fractionally. The common comments of almost all new riders is how often they have near misses that are not their fault. While I think Bravus's "nothing to report thread" said it best that with experience you no longer are having notable near misses every ride, in fact most of us who have been riding for a while will go for months or years without displaying the slightest ire towards a car doing the wrong thing. When you know how to ride you know how to position yourself so most of the time you don't need to be seen, and when you do need to be seen you position yourself to best effect that. New riders are exceptionally bad at this so some form of visual enhancement could potentially be useful.

    That being said, a high vis vest as the only option? I have a bright orange bike, putting a high vis vest on me would be like the 1 goal Brazil scored against Germany this morning, completely useless.

    Anyway, I've got no time to look through the research properly ATM, sorry.
  3. I know everyone get up in arms about Hi Vis being forced upon us. I too don't want to be told what to wear and have never worn it myself. However I always hear from people vehemently opposed to it things along the lines of "It doesn't make you more visible". I sure find it easier to see, and thousands of workers wear it without any uproar. Opponents also think that people who ride in Hi Vis automatically act they have a safety force field around them and forget about riding defensively which I also disagree with. As Mentioned in OP, it certainly can't hurt.
  4. Don't we mostly crash on our own outside of urban areas?
  5. I guess this mandating hi-viz could be seen as a test of its efficacy. So now we need to make sure that lots of learners in Victoria have accidents, negating the expected reduction in accidents. Over to the Victorians...

    But seriously, the mandating of hi-viz should give some clear indication as to whether it does change rider and driver behaviour for these new riders. But the research needs to be done.
  6. If you introduce the hi-viz at about the same time as improving rider training in the same part the country, what do you think will happen to accident rates for new riders?
    • Agree Agree x 1
  7. #7 smileedude, Jul 9, 2014
    Last edited: Jul 9, 2014
    Shouldn't they conduct a trial first? It's a pretty simple experiment, randomly select 2000 new riders, 1000 are the control and free to wear what they want and the other 1000 are paid to wear high vis. They have to remember that the treatment is not high vis makes a difference and but mandating high vis makes a difference. So if control riders choose to wear high vis it shouldn't make any difference to the experiment. Otherwise it would pose an ethical challenge as nobody would be able to tell riders they are not allowed to wear high vis. After a year see which group has had more multi vehicle accidents.
    Do this with L platers, P platers and fully licensed.

    Id hypothesize it would have a strong affect with L platers and a weak effect with red Ps. Null effect on anyone else.

    This couldn't be done on an observation study as the people who choose to wear high vis no doubt ride completely differently to those who don't.
    • Like Like x 3
  8. True. Trust politicians to do things that confuse the issue.
  9. I choose to wear a Hi Viz vest on the dumbass assumption it might help me survive longer by being that much more noticeable. I can also tell you that even with one on, dumbass car drivers still won't see you in round a bouts, because as previously stated, they simply are not looking for bikes. I just shake my head as I avoid them, and continue to wonder at their ignorance as I ride away.
    • Like Like x 1
  10. It might be possible to get an answer using existing data. You could use single vehicle accidents as a control. Compare the percentage of riders coming off wearing high vis in single accident vehicle accidents where police attended to the percentage of riders involved in multi- vehicle accidents that are wearing high vis and police attended. Assuming that high vis makes no difference to protection in a SVA than if there is a lower percentage of multi vehicle accidents with high vis than there is some evidence to support the laws. There is some confounding obviously and a fair few assumptions but this should account mostly for the "I'm a super duper safe rider crowd" who are wearing high vis and riding like vaginas. Anyway, IMO publishing these figures would be far better than the current study.
    • Agree Agree x 2
  11. Since this is in the Research, Studies and Data forum, it'd be good if we kept discussion to the studies and their strengths and failures.
    Your personal experience can be related in other hi vis threads.
  12. unfortunately I think the studies and the evidence are going to be almost irrelevant in this case. "it's hi-viz so it must be safer" is always going to win regardless of the data.
  13. My concern is that all these studies seem to start from the point that the process of looking (by drivers) is inevitably passive and beyond management.
    Not so. A behavioural optometrist of my acquaintance tells me that a relatively simple program of formalised training (as opposed to traditional self-interpreted education) would yield far better results in terms of drivers acquiring a life-long habit of 'looking to see'.

    It is true to say that some things will be incidentally seen more readily than others, but only if the viewer's brain perceives them as a threat. A short period of repetitive training can significantly enhance that response. It need not be expensive or time consuming.

    If the studies recommend measures that rely solely on what drivers incidentally see, then the greatest opportunities have already been lost.
    • Agree Agree x 2
    • Like Like x 1
  14. I think this approach would be biased in the other direction. I'm not sure how many hi-vis wearers get out and about in the hills. You might need to look at "single vehicle urban accidents" and see what contribution the wearing of hi-vis has to your risk.

    Personally I am the sort of person who would happily wear a free hi-vis jacket but the one I bought without it was cheap -- do I still get the risk reduction from being the sort of person who would choose to wear it?

    The OP has an underlying assumption that the government wants to do what is best for motorcyclists. I can't see how that is true, but it is nice thought!
  15. #15 robsalvv, Jul 9, 2014
    Last edited: Jul 9, 2014

    From the SWOV study:

    Intuitively it seems to make sense to increase conspicuity of motorcycles by increasing their physical appearance. There are indications that physical appearance has an effect on crash risk. A large population based case-control study in New Zealand found that increased conspicuity of motorcycle and rider reduced the risk of motorcycle crashes with severe or fatal injury (Wells et al., 2004a). Drivers wearing reflective or fluorescent clothing had a 37% lower risk of crash-related injury, wearing a white helmet was associated with a 24% lower risk (compared to a black helmet), and voluntary use of DRL was associated with a 27% lower risk of crash-related injury. Of course, with a case-control study there is a risk of confounding factors, e.g. riders wearing highly visible clothing and/or helmet are likely to be more safety conscious than other riders. Therefore, it is not clear if the same crash reduction could be achieved for every motorcycle rider who, for example, is obliged to wear a white helmet.

    In contrast, there are also studies indicating that in some environments wearing fluorescent clothing did not improve conspicuity. Research indicates that the most important aspect of motorcycle conspicuity is contrast with the environment.

    = = = =

    The Wells study was based on data from the 1990's. I'd say that in those days, a rider choosing to wear bright coloured conspicuous gear for their safety, was most likely an inherently conservative rider and as a result YOU WOULD EXPECT conservative riders to feature less in crash data. So the confound is very clear - was it the conspicuity or the riding style that reduced their presence in crash stats? Which had the biggest influence on the result?

    This is the danger of cherry picking research.

    The current proposal for Hi Viz is that the Hi Viz must comply with AS4602 High visibility safety garments - Garments for high risk applications. That means that no current motorcycle gear that includes hi viz panels or is indeed a hi viz colour complies. Not every motorcycling scenario is a rider on a conventional bike going on a discretionary ride. There are some 17market segments and many many different reasons and purposes for why someone will swing a leg over a bike at any time... this simplistic "solution" for an ill defined problem will come up against difficulties.
  16. You have to remember that Vic Roads is a government department. Common sense and logic have no place there. They are typically bureaucrats, so they work like Sir Humphrey Abbleby (Nugel Hawthorn) so beautifuly put it:

    1. We have a problem
    2. We must do something
    3 This is somehting (Hi Vis Solution)
    4. Therfore we Must do this

    In other words, it's more important to be seen to be doing something about the problem than actually solving the problem.
    • Like Like x 1
    • Agree Agree x 1
  17. So, there is every possibility that a learner rider might choose to forgo motorcycle-specific protective gear in favour of AS4602 compliant garments, believing they are both legally required and safer?
  18. Good question, I don't know, but it will be a case of wearing something AS4602 compliant over the top of anything else you are wearing.
  19. Is there an Aussie post report on the number of crashes before and after they introduced mandatory Hi Vis for their Bike riders?
    Anecdotal evidence, ie - talking to posties riding bikes, seems to be it's made bugger all difference. They still get run over.
  20. I've only seen the submission to the PIMS which made claims about the injury reduction since new hi viz was introduced. But it's far from a scientific report, you can't tell whether other factors have had an influence, e.g, the simultaneous introduction of a new safety program addressing other underlying issues.

    Anecdotally, the posties I've spoken to say it hasn't made much difference, but in the new system they'd be reluctant to report minor incidents due to the overhead involved.