|
|
Thread Tools | Display Modes |
#41
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Monday, March 20, 2017 at 11:06:06 PM UTC-7, Jeff Liebermann wrote:
On Mon, 20 Mar 2017 14:26:35 -0700, sms wrote: Anytime someone doesn't like the results of a study they try to pick it apart. Actually, common practice is to first blame someone and then pick apart the argument. However, I prefer to undermine the study and let it collapse under its own weight. If "pick it apart" is an unacceptable method of debating the merits of a study, what would you consider to be an acceptable method for this newsgroup? I could use propaganda, various logical fallacies, anecdotal evidence, my personal feelings, or perhaps fabricate a contradictory study. Methinks that "pick it apart" is the same as breaking down the study into individual claims and seeing how each one holds together under stress. Perhaps it would be helpful if I explained how I analyze such studies. I've done it in this newsgroup at least a dozen times, but have never really explained how it's done. First, I find the original study. This is the most difficult part because studies are now hidden behind pay walls, revised continuously, and "edited for publication" in different lengths and forms. Once I have the original study, I try to determine who paid for it. That's because the conclusions and summary of the study are owned by whomever paid for the study, while the actual data and calculations are owned by the academics, scientists, students, and statisticians that ran the study. Often these are different or even in opposition. I then read the study in as much detail as I have time available. That's when the differences between the study and the web page announcing the study become apparent. In medical studies and surveys, I've seen claims that are quite the opposite of what the research shows, usually because the claims support a product or remedy. From this point, my approach varies depending on what I'm trying to demonstrate, prove, denounce, or evaluate. Usually, pointing out inconsistencies, gross omissions, and occasionally math errors is sufficient. In this case, I have been unable to find the study in either the original Danish or an English translation. Therefore, I have not read the original and have had to work with a brief summary from some unknown report or survey that apparently has been quoted and recycled extensively. The best I could do is point out that the percentage cited was meaningless without also disclosing the statistical population (number of participants in the test). This is hardly "pick it apart". So, I'll pick at it some more. One problem with claiming that flashing tail lights reduce accidents is that there just might not be any correlation between tail lights and accidents at all. Just because two things correlate (follow the same trends) does not mean that one causes the other. Some ludicrous examples: http://www.tylervigen.com/spurious-correlations So, how does one prove that flashing tail lights actually cause a reduction in accidents and that the 30% drop was not a coincidence? Well, one way is play the record backwards. Instead of giving out tail lights, find a group that has been using flashing tail lights for some time and take away their tail lights. If accidents increase, then there just might be a connection. Perhaps programming the tail lights so that they flash at different rates under the assumption that a faster flashing rate is more visible and therefore safer. I could dream up a few more tests, but basically the idea is to do things that test for a connection between flashing tail lights and accidents. The other part of the problem is that it's very easy to demonstrate that something is unsafe. All that's needed is one accident. However, it's impossible prove that anything is safe because there will always be accidents caused by coincidence or disconnected correlations. Have I "picked apart" your one liner sufficiently? I have never believed that a small barely visible flashing light on the back of a bike could possibly be more visible that a bright red and yellow jersey. How could it compare to a dayglow green or yellow jacket? The very idea that a large bike with a large person on it would be invisible unless protected by an LED the size of a pen-head with a lens on top of it to spread the light to the point where it's visible is pretty rediculous. |
Ads |
#42
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Monday, March 20, 2017 at 11:23:35 PM UTC-7, Jeff Liebermann wrote:
I see one problem with the study. There's little correlation between accident crash rate and being conspicuous. The problem that drives of vehicles that hit bicyclists almost always proclaim that they didn't see the bicyclist. That might be because the bicyclist was not easily visible, but could also be because the driver wasn't paying attention, was distracted, in desperate need of corrective vision, or was under the influence of booze, drugs, or passengers. For these drivers no amount of conspicuous clothing or flashing lights will improve their driving. Let's remember that the most common excuse for a car accident is, "I didn't see him." So why should it be any different for a car-bicycle crash with equal invalidity? That begs the question of what is the ratio of attentive drivers to impaired losers? I don't know. If I arbitrarily assign a 50/50 distribution, then I'll probably find that the overwhelming majority of bicycle crashes are caused by the impaired losers. That means that visibility has little effect on the conscientious drivers, who will probably be paying attention to their driving, and little effect on the impaired losers, who will probably be immune to any improvements in visibility. I was pulled over by a cop and given a mechanical ticket (out of date tag). How did he pick me out of the lot of San Mateo Bridge traffic? I was the only one driving the speed limit and attentive. This says an awful lot about the attentiveness of the average driver. |
#44
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights forBicycles.
|
#45
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Tuesday, March 21, 2017 at 6:23:35 AM UTC, Jeff Liebermann wrote:
On Mon, 20 Mar 2017 16:13:39 -0400, Frank Krygowski wrote: On 3/20/2017 9:59 AM, jbeattie wrote: Check this out: https://academic.oup.com/eurpub/arti...enting-bicycle Don't ride in Auckland, even with a blinky. Interesting study, with weird results. Seemed the group that was just "occasionally conspicuous" had the lowest crash rate. (...) I see one problem with the study. There's little correlation between accident crash rate and being conspicuous. The problem that drives of vehicles that hit bicyclists almost always proclaim that they didn't see the bicyclist. That might be because the bicyclist was not easily visible, but could also be because the driver wasn't paying attention, was distracted, in desperate need of corrective vision, or was under the influence of booze, drugs, or passengers. For these drivers no amount of conspicuous clothing or flashing lights will improve their driving. That begs the question of what is the ratio of attentive drivers to impaired losers? I don't know. If I arbitrarily assign a 50/50 distribution, then I'll probably find that the overwhelming majority of bicycle crashes are caused by the impaired losers. That means that visibility has little effect on the conscientious drivers, who will probably be paying attention to their driving, and little effect on the impaired losers, who will probably be immune to any improvements in visibility. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 Is there any research that breaks down conspicuity (`for instance bright clothing) and "active" devices like flashing lamps? It seems to me that, without disagreeing fundamentally with your eminently sensible argument that accidents are caused by the self-impaired losers, that even losers would some of the time catch a strong enough blinky in their peripheral vision, jerk the wheel and slam on the brakes, and perhaps kill themselves before they kill a cyclist. Andre Jute Bring back eugenics for dangerous drivers! |
#46
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Tue, 21 Mar 2017 10:44:21 -0700 (PDT), Andre Jute
wrote: Is there any research that breaks down conspicuity (`for instance bright clothing) and "active" devices like flashing lamps? Probably, but I couldn't find much online that would be considered relevant. There's plenty about the effects of camouflage, which todays fashions seem to emulate. The problem here is that in order to obtain statistics on visibility and accidents, the clothing and illumination of the bicycle rider would need to be documented. To the best of my limited knowledge, this has not been done. In order to prove my point that inattentive losers are the drivers that most often cause car+bicycle accidents, the mental state of the driver would need to be determined. Since that is most often volunteered by the driver involved, the data would inconclusive. About all I can add to my contention is that of the 3 local bicycle riders that were killed on the highway, all died as a result of a drunk driver. It seems to me that, without disagreeing fundamentally with your eminently sensible argument that accidents are caused by the self-impaired losers, that even losers would some of the time catch a strong enough blinky in their peripheral vision, jerk the wheel and slam on the brakes, and perhaps kill themselves before they kill a cyclist. Based on my personal experience as a rather marginal driver, I would guess(tm) that the collision would be over well before the inattentive driver even realized that he had just run over or hit a cyclist. The local bums and transient tend to ride their bicycles without much concern about getting hit by vehicles. The bums expect car drivers to do something radical in order to accommodate their erratic riding style. This has happened to me several times while simply exiting the driveway from my office at very low speeds. I would come very close to hitting one of these black clad and unlit cyclists, which required me slamming on my car brakes and coming to a panic stop. If the timing had been only slightly different, I would have hit the rider, and only then realized what had happened. If that's the case (as I contend) with many accidents, enhanced lighting or clothing would do nothing because the driver literally has not seen the cyclist BEFORE the accident. However, your contention that an inattentive driver might be brought out of their slumber by the use of attention getting devices such as flashing lights, flare guns, loud noises, concussion grenades, etc seems possible for some circumstances. If the driver is drugged, drunk, or otherwise running on reduced capacity, I would guess that the accident would be over before they respond to such stimuli. If the driver is sleepy, then response time is slowed, but at least there's a chance of getting their attention. If they're distracted by a cell phone conversation, it takes time for them to change context and analyze the situation. In other words, I think it's possible, but unlikely for a driver to react sufficiently quickly. But, what about riding down the road with a flashing light that could be seen for miles? Wouldn't that get a drivers attention long before they need to react? I don't think so. People tend to follow set patterns. In this case, drivers know what to expect on the road. For example, I got hit by a dentist in his Pontiac while I was riding on the wrong side of the road. He was operating at full capacity, yet didn't see me because he didn't expect to see anything in the wrong way bike lane. It's much the same with blinking lights on bicycles. Drivers do not expect to see these which causes the driver to make them disappear until it's too late. At best, they might take extra time to identify this unusual and strange flashing apparition, which only increases their reaction time. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
#47
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Tue, 21 Mar 2017 10:21:55 -0700, sms
wrote: On 3/21/2017 7:31 AM, wrote: On Monday, March 20, 2017 at 7:42:32 PM UTC-7, John B. wrote: Unfortunately it does not conceal the fact that what you stated, "a comparison of bike lights versus no bike lights", was not what the Odense study tested, nor was it the results of the study. But John, the whole point is that you have NO IDEA what they accomplished with a study that so obviously had such a small study group that they wouldn't even publish the size of it. You know that in statistical analysis concerning small percentages of injuries and fatalities as bicycle accidents that the study size has to be gigantic to reveal any pertinent information. So why would you pretend differently? You do realize how statistical sampling works don't you? The study had 4000 participants, 2000 with the lights, 2000 without the lights https://books.google.com.au/books?id=LvthAgAAQBAJ&pg=PA168. This is an _enormous_ sample for a country of that size. Denmark has about 4 million residents of cycling age. About 55%, or 2.2 million, cycle. The study had 4000 participants, 2000 with lights and 2000 without lights. This would produce a result with a 2% margin of error and a 99% confidence level. Even if 100% of those of cycling age cycled, the sample size needed barely goes up. Whatever criticism you may have of that particular study, sample size cannot be one of them! Nope. The facts a According to the Bicycling Embassy of Denmark, 9 out of 10 Danes own bicycles, while only 40% own cars. 17% of all trips are by bicycle and 20% of all commuters travel by bicycle, on the average, Danes cycle 1.5 km a day, 44 % of all children aged 10-16 cycle to school. The population of Odense is some 175,245 and: 31% of all people visiting the city center of Odense arrive by bicycle. Bicycle traffic constitutes 24% of all traffic. Odense has more than 545 km cycle lanes. There are some 26,000 students in Odense University, and if the same parentage of collage students cycle as does the 10 - 16 age group there are 11,000 cyclists in the University, of whom, apparently, some 36% participated in the study. In short a country where bicycling is not only an accepted form of transportation but a very, very, commonly used form of travel. Given that from what I read Reelight gave a set of lights to each of the 2,000, "with permanently on light" participants in the study I don't find the numbers to be especially astonishing. Another point. All bicycles in Denmark ridden after dark must, by law, be equipped with a white headlight and a red rear light. Failure to comply with this law results in a 700 Kroner fine. Given the heavy use in Denmark I would suggest that all, or nearly all bicycles are equipped with lights. As far as I can tell 700 DKK is about 20% of the average monthly salary. This would produce a result with a 2% margin of error and a 99% confidence level. Even if 100% of those of cycling age cycled, the sample size needed barely goes up. Whatever criticism you may have of that particular study, sample size cannot be one of them! I see nothing wrong with the study. The objection was the definition of it as a study of bicycles with lights and those without lights. Which was.... just not true at all. Now, I see, the story has been changed and the lights/no lights story has morphed into a whole new story. Very adroit footwork I must say. -- Cheers, John B. |
#48
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights for Bicycles.
On Mon, 20 Mar 2017 23:37:08 -0700, sms
wrote: On 3/20/2017 11:06 PM, Jeff Liebermann wrote: On Mon, 20 Mar 2017 14:26:35 -0700, sms wrote: Anytime someone doesn't like the results of a study they try to pick it apart. Actually, common practice is to first blame someone and then pick apart the argument. However, I prefer to undermine the study and let it collapse under its own weight. If "pick it apart" is an unacceptable method of debating the merits of a study, what would you consider to be an acceptable method for this newsgroup? I could use propaganda, various logical fallacies, anecdotal evidence, my personal feelings, or perhaps fabricate a contradictory study. Methinks that "pick it apart" is the same as breaking down the study into individual claims and seeing how each one holds together under stress. There is a tendency to nitpick little things and then to declare the entire study as worthless, when in fact, other than perhaps in drug trials, there is just not going to be a "perfect study." Yet the goal of the study was to determine if flashing lights were effective, and if so, use the data to remove a ban on flashing lights. The company that was involved in the study certainly had a vested interest in the outcome, but they are only one of a multitude of companies that are benefiting from the outcome. (...) You ignored my question. If "pick it apart" is an unacceptable method of discussing the merits of a study, what is an acceptable method? Picking at the details while ignoring the main points is common enough. You are doing it right here in this discussion. My main point was that there is no proof that a correlation between bicycle crashes and flashing lights constitutes causation. You ignored that and went on to deal with the trivia. Let me propose a method, extracted from my previous rant, which I use: 1. Go to the original source. Avoid summary or survey sources. 2. Find what person or organization is paying for the study. That often reveals a hidden agenda. 3. Compare the abstract and summery in the report with the actual data. Often, they're quite different. 4. Look for inconsistencies, dubious sources, "normalization", pre-selection of participants, and statistical creativity. All this is certainly "pick it apart" methodology. Oddly, I can't find a better way to deal with a study or report. You also seem to ignoring the not so trivial problem that we *ALL* are discussing the issue without access to the original report and numbers from the trial. I would like to see the number of accidents with and without the flashing lights, the methodology, and how the numbers were produced before blundering further with my guesswork. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
#49
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights forBicycles.
On 3/21/2017 10:27 AM, AMuzi wrote:
On 3/21/2017 1:06 AM, Jeff Liebermann wrote: On Mon, 20 Mar 2017 14:26:35 -0700, sms wrote: Anytime someone doesn't like the results of a study they try to pick it apart. Actually, common practice is to first blame someone and then pick apart the argument. However, I prefer to undermine the study and let it collapse under its own weight. If "pick it apart" is an unacceptable method of debating the merits of a study, what would you consider to be an acceptable method for this newsgroup? I could use propaganda, various logical fallacies, anecdotal evidence, my personal feelings, or perhaps fabricate a contradictory study. Methinks that "pick it apart" is the same as breaking down the study into individual claims and seeing how each one holds together under stress. Perhaps it would be helpful if I explained how I analyze such studies. I've done it in this newsgroup at least a dozen times, but have never really explained how it's done. First, I find the original study. This is the most difficult part because studies are now hidden behind pay walls, revised continuously, and "edited for publication" in different lengths and forms. Once I have the original study, I try to determine who paid for it. That's because the conclusions and summary of the study are owned by whomever paid for the study, while the actual data and calculations are owned by the academics, scientists, students, and statisticians that ran the study. Often these are different or even in opposition. I then read the study in as much detail as I have time available. That's when the differences between the study and the web page announcing the study become apparent. In medical studies and surveys, I've seen claims that are quite the opposite of what the research shows, usually because the claims support a product or remedy. From this point, my approach varies depending on what I'm trying to demonstrate, prove, denounce, or evaluate. Usually, pointing out inconsistencies, gross omissions, and occasionally math errors is sufficient. In this case, I have been unable to find the study in either the original Danish or an English translation. Therefore, I have not read the original and have had to work with a brief summary from some unknown report or survey that apparently has been quoted and recycled extensively. The best I could do is point out that the percentage cited was meaningless without also disclosing the statistical population (number of participants in the test). This is hardly "pick it apart". So, I'll pick at it some more. One problem with claiming that flashing tail lights reduce accidents is that there just might not be any correlation between tail lights and accidents at all. Just because two things correlate (follow the same trends) does not mean that one causes the other. Some ludicrous examples: http://www.tylervigen.com/spurious-correlations So, how does one prove that flashing tail lights actually cause a reduction in accidents and that the 30% drop was not a coincidence? Well, one way is play the record backwards. Instead of giving out tail lights, find a group that has been using flashing tail lights for some time and take away their tail lights. If accidents increase, then there just might be a connection. Perhaps programming the tail lights so that they flash at different rates under the assumption that a faster flashing rate is more visible and therefore safer. I could dream up a few more tests, but basically the idea is to do things that test for a connection between flashing tail lights and accidents. The other part of the problem is that it's very easy to demonstrate that something is unsafe. All that's needed is one accident. However, it's impossible prove that anything is safe because there will always be accidents caused by coincidence or disconnected correlations. Have I "picked apart" your one liner sufficiently? I don't know but significance and meaning may vary. Just read a newspaper headline about a "new drug found 70% better than aspirin". Of 3600 people over 5 years there were 2.3% heart attacks in the daily aspirin group and 1.6% heart attacks in the new new group. Good example. I've mentioned this before, but that specific point is treated in great detail in the book _Bad Science_ by Ben Goldacre. He says both researchers pushing either the drug or their own publication invariably use the "70% better" line; but most medical issues (and, I'll add, "bike safety" issues) involve rare events. It's MUCH better to give the actual percentages. Or even better, what Goldacre calls the "natural frequencies" - out of 10,000 people, how many will get heart attacks while using aspirin? How many while using the new drug? In Scharf's case: Out of 10,000 bike commuters, how many will be hit while using no daytime lights? How many while using the advertised light that was given away to promote the study? The difference is probably very, very small. -- - Frank Krygowski |
#50
|
|||
|
|||
The University of Aalborg Study on Daytime Flashing Lights forBicycles.
On 3/21/2017 11:25 AM, sms wrote:
On 3/21/2017 7:24 AM, wrote: If they did not want it picked apart they only had to provide the actual numbers. And they didn't. Why do you suppose that was? The numbers are almost certainly there--if you pay for the full study. A lot of studies are like that. They publish a summary for free, but you have to pay for the full study. I guess that the thought is that it would be organizations with a budget for which a few hundred dollars (or in this case about $40) would not be a big deal. But in countries where flashing lights are already legal and widely used, and the benefits well-established, why would anyone pay anything just to get the raw data? Yesterday it was cloudy here. I was driving in the morning. Gray cars in gray conditions don't stand out. But you see cyclists with DRLs coming a mile away (literally), long before you see any bright clothing. I doubt if anyone here really believes that on bicycles DRLs (flashing or steady) are not effective. Just look at motorcycles which have been required to have a DRL for decades (at least in most states). The _very_ significant difference between bikes and motorcycles is the typical speed. The closing speed between a motorcycle and an oncomoing car can easily be 120 mph. That closing speed is vanishingly rare for a cyclist. What that means is that in situations where there actually is some risk of colliding - say, a left cross situation - the motorcyclist can be hundreds of yards away and thus almost invisible. The bicyclist's body and bike are much closer and much more visible than the motorcyclist's. It's true that gutter riders are less conspicuous in left cross and pull-out situations. But that should be solvable by cyclist education. Those (like me) who typically ride in more visible lane positions just don't have the dozens of close calls you frequently allude to. (I had precisely one, in about 1977, and it taught me to stay out of the gutter.) BTW, my wife and I made one of our frequent long drives home today. On a two lane highway, I saw a cyclist coming toward us with a front blinking DRL. Even if I did notice the light before the cyclist himself (I doubt it, but it's hard to say), I thought "What's the point?" He was in a bike lane on the other side of the road in a place with no intersections for about a mile; and I would have seen him easily if I did have to turn across his path, just as I've seen hundreds of daytime cyclists without DRLs in the past. Of course, he may have had a St. Christopher's medal. That may have been what really made me notice him. ;-) -- - Frank Krygowski |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
New B&M Lights with Wide Beam and Daytime Mode Available | sms | Techniques | 74 | November 23rd 13 07:46 PM |
Flashing front light in daytime. | Nick[_4_] | UK | 22 | June 18th 08 12:28 PM |
Car daytime running lights | David Hansen | UK | 15 | March 25th 08 12:55 PM |
Daytime running lights | Martin Dann[_2_] | UK | 7 | February 13th 08 12:15 AM |
New University of Illinois Helmet Study ... | Edward Dolan | Recumbent Biking | 1 | May 8th 06 02:04 AM |