Road Safety News

‘Failed to look’ is number one contributory factor

Monday 6th July 2015

The IAM says a Freedom of Information request to the DfT confirmed ‘failure to look properly’ as the most common contributory factor which is included in more than 30,000 collisions annually.

While police can record up to six contributory factors from a list of 77, the IAM says “the top two give the most obvious reasons for the incident”. 

Analysis of the 2013 contributory factors showed ‘failure to look properly’ and ‘failure to judge another person's path or speed’ as the top combination, responsible for 13,299 collisions, or 7% of the total.

Next was ‘failure to look properly’ combined with ‘carelessness or recklessness’, or ‘judged to be in a hurry’. These totalled 9,132 incidents, 5% of the total.

The full table can be downloaded here.

Sarah Sillars, IAM chief executive officer, said: “These figures show conclusively that simple human errors continue to cause the majority of accidents. Drivers cannot blame something or someone else for a collision happening, it is down to every one of us to make a difference.

“We feel that many people eventually get complacent behind the wheel and inattention creeps in. Combine this with fatigue and distractions, inside and outside the vehicle, and the message is clear that drivers must apply their full attention to driving – you simply cannot do two things at once if one of them is driving.

“We have consistently advocated that continuous assessment is one of the main ways to ensure no driver gets into bad behaviours that cannot then be rectified.”


Comment on this story
Report a reader comment

What's your view - comment on this story:

I confirm that I have read and accept the moderation policy and house rules relating to comments posted on this website.
Your comment:
Your name and location:
Your email:

Ha Ha! Hoist on my own petard I think!

Although you cannot know how stressful or pressured (or otherwise) was the environment within which I drafted that reply. But no excuse, I read one thing and typed another, should have triple checked it and didn't, my apologies. Over to you please, Andrew!
Honor Byford, Chair, Road Safety GB

Agree (0) | Disagree (0)

Taking proper and effective observation is vitally important when driving as the eyes are the most important of the senses. Looking and seeing are two different things. Many drivers look but don't see. Seeing is when the looking in transmitted to the brain to assess what is seen or not as the case may be. It is quite true that many drivers emerging from minor roads do not look/see properly as their 'zone of vision is limited'. They make their decision to emerge BEFORE the can see fully to the right and left. There is also the judgement issue of speed, distance and timing.
Tom Harrington LL B, Co. Kerry, S. Ireland.

Agree (4) | Disagree (0)

In collision reporting, would it not be better for the default 'cause' to simply be 'careless, reckless or in a hurry' anyway, instead of trying to break it down in to sub-categories of this overiding condition? The exception might be where there is some known, noteworthy factor, such as sudden illness at the wheel, or othe Act of God, evidence of impairment die to drugs/alcohol. As Andrew said earlier 'failing to look' or 'misjudgment' is rather stating the obvious and would probably cover most everyday collsions anyway.

As I have said on similar related news items, without knowing exactly what happened for each collison where these factors were entered, even if correct, it's not as informative as the IAM may like to think.
Hugh Jones, Cheshire

Agree (0) | Disagree (6)

I know I am normally the villain of the piece Honor, but I don't think I've queried the statistical collection method in my recent posts!

The misattribution of Andrew's post to me is an example of the failure mode we are discussing which is 'failing to look properly' and it shows how easily it happens even in non-stressful situations.
Duncan MacKillop. No surprise - No accident.

Agree (7) | Disagree (1)

I am concerned by your assertion that police officers attending a reportable collision just “ask a few questions” because an admin team sends out a questionnaire after the event. I would be grateful if you would provide me with more details please as to where this happens and in what cases – is this for damage only incidents, a response to those reported after the event by individual or across the board for investigated personal injury collisions?

Our understanding is that police officers attending/investigating a reportable collision will complete a Stats 21 form. The detail of the information that has to be collected in this form is agreed between the police, Home Office, Department for Transport and other stakeholders, including Local Highway Authorities who use the data to inform their engineering and educational work. The fields in Stats 21 forms are reviewed every five years in a process to which all parties contribute. This data is also used as part of the planning process and for transport modelling and academic research. If what you allege is true, we will look into it further.
Honor Byford, Chair, Road Safety GB

Agree (4) | Disagree (0)

Abnormal circumstances usually occur in rare and unique combinations which is what makes them so difficult to detect after the event. Abnormal circumstances then can be the result of a number of normal circumstances combining in a very slightly different way to normal. Take the SMIDSY as a prime example of the combination of a camouflaged motorcycle (perceptual or physical) AND an affordance available to a waiting car happening at exactly the 'right' time to cause a collision. The rare combination of these quite normal circumstances and not the circumstances themselves is the abnormality that is present. If the rider or driver cannot spot the abnormality they will continue to do what they would normally do which in the driver's case is pull out and in the riders's case is carry on without taking any particular steps to ameliorate the situation.
Duncan MacKillop. No surprise - No accident.

Agree (5) | Disagree (1)

I totally agree Derek that the Abnormal part is where a party, say the second, considered innocent party fails to assess the situation with a degree of objectivity and makes assumptions or decisions about what the first party is going to do and gets it wrong. The first party doing something the second party would not think about doing. The assumption being that the other or first or guilty party has the same notion of what is right and what is not.

Duncan's argument regarding bends relies upon the fact that they lead in on a large radius and then get tighter to end with a smaller radius. He describes this as an engineering fault and was initially used by engineers for the safe rounding of bends by TRAINS which needed to be assisted to slow on the approach to a bend that would otherwise derail them. I understand his point but they are only about 1% or less of all bends. This engineering was used by road construction and many can be seen on and off motorways as reducing radial bends. However on bends in Wales for an example it is the topography which dictates the sharpness of bends and many are reducing radius without any slowing lead in. What we need is a simple message on the approach to such a bend of its decreasing radius and obviously increasing danger. One which is easily identified and would inform all oncoming traffic of this particular danger.
Bob Craven Lancs Space is Safe Campaigner

Agree (4) | Disagree (1)

I would like to put a different perspective on the publication of this recent data.

There are 2 unrelated issues with it's use and interpretation. Firstly, the IAM, a well known and established advocate of advanced motoring principles which heavily leans towards improving observational skills, has in recent years been raising it's game. It is more and more interested in increasing it's financial gains with marketing rather than what it does for drivers and riders. So I'm a little suspicious that the IAM will use the headline cause of collisions 'fail to look properly' as another furtherance to sell the cure for failing to look properly by advanced training with them.

The second issue, is we seem to accept this data without question. Not questioning the truth of the number of times that 'failing to look properly' was recorded by police as the causation factor, but shouldn't we question whether the use of that cause code was correct.

The current police practice now is to attend one incident and move on as quickly as possible to the next. I suspect that the police now, with all good intention, turn up at a collision, ask a few questions (only a few because an admin department send out pro-forma questionnaires for answers to the other questions which should have been asked at the scene) and decide for themselves what cause code they will record. Assuming very, very few people intentionally pull out into the path of another vehicle, it isn't suprising that failing to look properly is the easiest cause to attribute to collisions when vehicles come together.

What the officers could and should be delving into is why the person failed to look 'properly'. It could be many reasons which in truth would change the true position and meaning of 'fail to look properly'.
Andrew, Lincs

Agree (7) | Disagree (2)

If I may;

I see Duncan's abnormal circumstance as being one in which is the opposite to normal. For example, many bends may be taken safely which is the norm. Then one is taken too fast, control is compromised, and a crash occurs. This is the abnormal.

Similarly at a junction. Many junctions can be negotiated safely, but then something out of the ordinary happens such as a vehicle (vehicle 1) emerging into moving traffic without notice, and a vehicle (2) in that flow of traffic is then subjected to an abnormal event. This in turn surprises the driver of vehicle 2, who fails to stop in time due to the situation being outside of what has been experienced as normal, and a collision occurs. Thereby the causation of the completing collision is shared between two parties.
Derek Reynolds, Salop.

Agree (5) | Disagree (0)

I am having difficulty in getting my head around what is an abnormal circumstance. Duncan, the two you have outlined seem to be approaching a junction and entering a bend. Both as being abnormal and yet I and many would consider those two actions to be absolutely normal.

What I believe is that once again as with a smidsy there is perhaps a lack of appreciation of the situation and the possible assumption that all is going to be well. That is if we, in the first instance the smidsy, presume that the other party will act in a certain way and therefore make no preparation for any alternative actions and end up in a collision. Similar could be said of a bend which is not assessed correctly for many reasons and one ends up getting into trouble. Again the attitude and cognative behaviour of the driver or rider has made a wrong assessment or prediction.
Bob craven Lancs.... Space is Safe Campaigner

Agree (1) | Disagree (0)

It's a common mistake to think that errors have to be co-located in time and space in order for the two to tangle rule to come into effect. The spiral easement to a corner is the result of a significant error that was made over 100 years ago yet it's still killing people with monotonous regularity. A rider encountering one of these horrible and deadly things can easily be fooled into thinking they have got their entry speed roughly right whilst in fact they have got it exactly wrong. The resulting crash is always blamed on the rider yet their true error was in not identifying the presence of the easement which represents the abnormal circumstances of which I spoke earlier.
Duncan MacKillop. No surprise - No accident.

Agree (3) | Disagree (4)

You're right in the context of these 'failed to look' junction collisions, but what about single vehicle accidents e.g. a bike or car taking a corner too fast and hitting a wall? This does not involve another party, so there can't be another party making a 'completeing error'. Similarly, a rear-end shunt where the victim was a sitting duck and was not making any completing errors.
Hugh Jones, Cheshire

Agree (3) | Disagree (2)

Vision, perception and their various failure modes are a big and very complicated subject and that makes them very difficult to learn and understand. What is easy to learn however is how to identify those situations where visual and perceptual failures are more likely to be made so that they can be more easily avoided. As most accidents happen when people do normal things in the presence of abnormal circumstances it is the identification of these abnormal circumstances that is key to collision avoidance.

We now know that it requires a precipitating error on behalf of one party in a collision to be compounded by a completing error made by the other party (it takes two to tangle) in order for a collision to occur. In the case of SMIDSY's the completing error by the rider is invariably their failure to identify the presence of abnormal circumstances and so by concentrating on helping people identify abnormal circumstances we will get the most bang for our buck.
Duncan MacKillop. No surprise - No accident.

Agree (4) | Disagree (2)

IMO this is not a very useful code. Self-evident really. How do you treat it?
Roger Harding, Warwick

Agree (2) | Disagree (0)

They don't. If you don't look you don't see. Having dipped headlights on motorcycles for many years, at least since the 1970s, and also the wearing of day glo jackets this century and improved extra lighting may be considered an advantage but even the police in all their regalia still suffer smidsys. Also it's not just in the domain of two wheeled vehicles, it happens frequently to car drivers also.

There is a train of thought coming from the No Surprise No Accident Campaign that the recipient or other vehicle makes a presumption that the offending vehicle will not pull out and that is the wrong presumption. By anticipating and presuming that it will, worse case scenario, the innocent driver or rider approaching has to make certain different decisions on how he or she would proceed to avoid a possible conflict with the offending vehicle.

No Surprise then when it pulls out but No Accident due to a correct presumption. By slowing, giving distance (sufficient to stop if needs be), moving away, giving distance away from a possible collision and showing oneself, and possible use of the horn a collision may be prevented or an injury mitigated.

Presume otherwise and end up in hospital when with the right defensive attitude it could be avoided.
Bob Craven Lancs..Space is Safe Campaigner.

Agree (7) | Disagree (0)

So how, in the light of this statement, do daytime driving lights on cars help motorcyclists be as visible as possible?
Doug Harris, Stockton-on-Tees

Agree (5) | Disagree (0)

Also Bob, not uncommon is drivers - especially those in a hurry - emerging to turn left who only look right but not left, in the mistaken belief that there won't be any vehicles on that side of the road - when sometimes there is - passing parked vehicles or even overtaking.
Hugh Jones, Cheshire

Agree (10) | Disagree (0)

Back to subject. There is a presumption that drivers have not been seeing properly other vehicles on the approach to a junction etc. Or having been seen the speed of the approaching vehicle has been calculated or time of arrival estimated incorrectly. However there are occasions when a driver does not fail to look properly, he or she just doesn't bother looking at all. And then there are those drivers who look but dont bother anyway when they see a motorcyclists or cyclist or even another car etc.

From my own observations I estimate that about 25 to 30% of drivers at a junction do not look to the right first but look the other way and then if clear pull out before they have looked to the right (if ever) at you approaching. Test this yourself and make a mental note when approaching a junction just how many drivers are looking away from you. You will be surprised.
Bob Craven Lancs...Space is Safe Campaigner.

Agree (5) | Disagree (2)

Perhaps that's not a bad thing as non professionals may be new thinkers and not restrained by others who for whatever reason cannot or do not contribute to this forum. If we were all of the same mind think what that would mean.
Bob Craven Lancs...Space is Safe Campaigner

Agree (9) | Disagree (2)

I worry for new readers to this site. They could be forgiven for believing the road safety profession is in complete disarray reading some of the comments. For the benefit of the casual rreader, I think it would be useful for them to be aware that not every contributor is necessarily from within or connected to the profession.
Hugh Jones, Cheshire

Agree (9) | Disagree (9)

HI Andrew

I think your analysis needs a little more explanation.

Rod King, 20's Plenty for Us

Agree (2) | Disagree (7)

Per Rod,

Stats are from Stats 19 2013 with contributory factors. Approximations can be derived from RRCGB 2013 by divding overall figures by the percentage of pedestrian injury overall, viz:

Cas: 7 / (24033/183670) = 53%

Fat: 11 / (398/1713) = 47%

Pedestrian factors are involved in (76% 76%) (casualties%, fatalities%)

Failing to look or judge (driver/rider/pedestrian) is (77 64)

Pedestrian factors, driver/rider error or behaviour is (95 94)

Pedestrians injure 3x as many non-pedestrians as speeding drivers injure pedestrians, and kill as many non-pedestrians as normal experienced (>30) speeding drivers kill pedestrians.

Pedestrians are dangerous and lethal to themselves and others. Perhaps someone should point that out to them, or heaven forbid, hold them accountable.

I do find it slightly disturbing that proponents of social control to 'protect' the pedestrians are unaware of or are willing to ignore the underlying data.
Andrew Mather, Kent

Agree (12) | Disagree (6)

Why did the IAM go to all the trouble of using FOI when the contributory factors for all years are already on the DfT website? I'm surprised the IAM didn't know that or find out simply by googling.
Dave Finney, Slough

Agree (21) | Disagree (2)

How does the reporting officer know that the driver(s) 'failed to look properly' anyway? Was that the drivers' version of events? "I just didn't see him" sounds better than "I was in a hurry and didn't really look properly". Guilty motorists do not necessarily come clean when accounting for their behaviour to the Police! They usually go for the least blameworthy excuse.
Hugh Jones, Cheshire

Agree (18) | Disagree (6)

Useful tip Steve: don't look down at your speedometer when there's a need to concentrate on the road ahead, anticipating surprise appearances by pedestrians etc. Conversely, if you must look at your speedometer, only do so when you're certain that you can afford to take your eye of the road ahead for the fraction of a second this will take. Same goes for checking mirrors, fuel gauge, heater settings, radio etc. Driving 'slowly' affords more time to notice and anticipate hazards anyway. In urban areas, it makes no difference whatsoever to journey times, as I think you said elsewhere.
Hugh Jones, Cheshire

Agree (12) | Disagree (5)

A 'failure to look properly' is clearly some form of error yet how does a person know if they are in the process of making the error in the first place? Usually errors are revealed to us after the event in the form of a surprise, but as far as we know there is no pre-event error trapping system in the brain that we can rely on to reveal errors before they cause problems.

Millions of years of evolution have taught the human brain that it is much more efficient for errors to be revealed after events rather than before them which on the face of it seems rather odd. When you consider however that the brain only learns from error then you will understand why errors are left to play-out rather than being trapped and eliminated. We carry this evolutionary solution with us every time we go out on the roads and for the most part the 'error - surprise - learning' loop serves us very well as it teaches us an enormous amount about how the world out there actually works. It does however leave us as a hostage to fortune when an untrapped error occurs, but with no time available to recover from the situation that it has made for us.

We can take as much care and pay as much attention as it's possible to do and yet these errors can still occur because they are not revealed to us before we make any go/no go decisions. The question we must ask ourselves therefore is how do we reduce the number of visual/perceptual errors when the only way we know we are making them is after they have occured and not before? Answering this question is going to be of critical importance if we want to reduce the number of failure to look incidents.
Duncan MacKillop. No surprise - No accident.

Agree (10) | Disagree (6)

I wonder if travelling at 20mph could actually be a contributing factor to the inattention and failing to look. I would propose it is the incorrect assumption of safety at uncomfortably slow speeds or avoiding and looking out for obstacles built in the carriageway that take the drivers attention away from what should be first and foremost. Other road users and the prevailing conditions. If I am looking at my speedometer due to fear of prosecution that's precious moments my eyes are not on the road.
Steve Armstrong, Halifax UK.

Agree (13) | Disagree (11)


Can you give the source for those statistics please? Thanks
Rod King, 20's Plenty for Us

Agree (6) | Disagree (3)

It's a pity the IAM don't make fuller use of their many FOI enquiries. Why not also ask if any prosecutions result from cases where 'carelessness and recklessness' are cited? I know some Police are inclined to think that collsions are 'just accidents', or 'one of those things' and are not inclined to prosecute, especially the less serious ones, however the causation factors are the same, it's just the outcome that varies. Even a slight injury collision is still somebody's fault.
Hugh Jones, Cheshire

Agree (8) | Disagree (5)

We have seen the results of 20mph speed limits on casualties, and there has been no significant reduction other than opinions, which vary greatly under the circumstances of each individual collision such as the hour of the day, the volume of traffic, the urgency of wanting to get from A to B, and weather. If - as is suspected, a 20mph limit sees no improvement, will 15, or 10mph be next? This would simply give all those pedestrians and other road users who wish to place themselves into a carriageway irresponsibly, a better chance of no injury. But will develop a society of deliberately careless individuals with no consideration for others, or their own safety.
Derek Reynolds, Salop.

Agree (13) | Disagree (8)

Of course it also fails to recognise that pedestrian failing to look properly is also the single biggest cause of pedestrian accidents, with a far greater percentage (62% casualties, 45% fatalities). Just as well we're all headed for twenty.
Andrew Mather, Kent

Agree (15) | Disagree (3)

This rather puts brains, vision, cognition, perception, attention, distraction, prediction and surprise at the top of the list of things to understand so we can work out their role in accident causation.

Ms Sillars is a bit wrong when she says that "simple human errors continue to cause the majority of accidents" because there is absolutely nothing simple about human error. It is an enormously complex field of knowledge, but we do need to fully understand it in order to make any progress. Otherwise the fact that the IAM have made this FOI request in the first place shows that perhaps we are seeing some new and most welcome thinking from the road-safety establishment.
Duncan MacKillop. No surprise - No accident.

Agree (12) | Disagree (3)