Here's the hazard perception test

[Count Steer]

If the hazard test is so unrelated to the real world they could extend the principle for the driving test and just ask you how good you are at Grand Theft Auto instead of faffing about wasting petrol.

[Scotsrich]

8/15

I thought I’d got most of them but it turns out you need to ignore possibly hazards and just click away at the most obvious one in each clip.

Sorry but that’s a crap test.

[Dickyboy]

Absolute failure, take my licence away now... 3/15

[Dodgy69]

One man's hazard, is another man's norm.

[Pirahna]

The hazard perception test in a driving test centre is a doddle, I did it for my truck test in 2020. Spot the hazard, click the mouse, count five seconds, click again as it develops, count five seconds and click again if it's still there. On the actual test the first one caught me out because although you're clicking a computer mouse there is no pointer on the screen to point at the hazzard, it's just a matter of clicking when one appears. I got 98%.

Theory test was similarly easy, I must have got one wrong as I only scored 96% on that.

[McNab]

On the one with us following HiViz man, where are we supposed to click for the hazards?

I clicked for the side road, again as the pickup approached the junction from the left. Then clicked for the van in lay-by and again as he pulled onto the road. Er and that was it, apart from clicking for a blind hill and later when I couldn't see the van's left indicator, and that was only 'cos nothing seemed to be happening and I thought I should click something.

I only got 4 points which was for the van in the lay-by.

Edit:

So, is it just the one hazard per video? Having now done all 3 I scraped through with 10 points. The pickup approaching the junction isn't a hazard despite a very similar situation to the milk van? That winding country road was full of hazards, the water in the road, tight bends, blind single track bridge but only the tractor was worthwhile? And I pegged the tractor waaaay earlier than the test wanted me to.

[Wossname]

See what I mean (other thread)? You're all experienced motorcyclists, and you can't get a decent score (apart from Skub - 12? fiddle!). It's training people NOT to react till it's too late. I think that's positively dangerous and a Bad Thing (@1066 and all that).

But its introduction was associated with an 11% drop in new driver crashes!

https://www.gov.uk/government/news/haza ... collisions.

Thinking about this....was anything else "new" introduced at around the same time which could have contributed to tha 11%? And...

It occurs to me that the HPT may provide new drivers with their first real opportunity to react to an unexpected hazard without being actively prompted to by their instructor. I can't imagine any driving instructor sitting mute while their pupil approached any of those 3 clip situations - it'll be "look out for that milkman - he might..."/ "You can see that tractor approaching our road - he might..."/ "That van might pull out of the layby..." It would take some self control - on my part, certainly, when I was taking our kids out for practice - to NOT react, and driving instructors are human. Sitting at a screen on your HPT, nobody's going to dig you in the ribs to "do" something, So it may be a help - but could be so much better.

[Dodgy69]

Were the wet roads and bridge classed as potential hazards. Not forgetting high viz man following too close and all them cars on driveways that might move.

[Horse]

But its introduction was associated with an 11% drop in new driver crashes!

https://www.gov.uk/government/news/haza ... collisions.

Thinking about this....was anything else "new" introduced at around the same time which could have contributed to tha 11%?

Not as far as I know, and this analysis involved something like 50,000 drivers.

It occurs to me that the HPT may provide new drivers with their first real opportunity to react to an unexpected hazard without being actively prompted to by their instructor.

Could be right, it's the best suggestion I've heard.

[McNab]

Were the wet roads and bridge classed as potential hazards. Not forgetting high viz man following too close and all them cars on driveways that might move.

No, there is only one hazard per video. If you review the results it shows where they wanted you to click. I guess clicking on hazards they don't care about doesn't alter your score.

[The Spin Doctor]

But its introduction was associated with an 11% drop in new driver crashes!

https://www.gov.uk/government/news/haza ... collisions.

Thinking about this....was anything else "new" introduced at around the same time which could have contributed to tha 11%?

Not as far as I know, and this analysis involved something like 50,000 drivers.

It occurs to me that the HPT may provide new drivers with their first real opportunity to react to an unexpected hazard without being actively prompted to by their instructor.

Could be right, it's the best suggestion I've heard.

Here's the possible - perhaps even likely - flaw in the study quoted:

"This is a major study of learner and new drivers in which ‘cohorts’ of candidates taking the practical driving test were invited to participate in a series of postal surveys.

So the study didn't look at 'before and after' across a random sample of riders, which would have offered a reasonably accurate picture of the effectiveness of the hazard perception test. Postal studies have been shown to be have significant levels of inaccuracy in medical research into surveys like this. It's partly because of inaccurate reporting.

One paper suggests: "Within that sample, certain people may be more likely to respond to a particular survey than others. This may result in selection of a subset of the targeted population being represented."

I can certainly see a situation in which drivers who HADN'T crashed were keener to respond than those who had.

A second paper on postal surveys suggests: "About 10% of responses were inaccurate".

If the study had actually dug down into the actual crash data and compared the crash rate of new drivers before and after the introduction of the Hazard Perception test, I would have more confidence.

I don't know if that data's easily available but an independent study for ALA Insurance in 2017: "From our research we found that just over one in five (21.6%) new drivers had been involved in an accident during their first year of driving."

One graph I did find...


Figure 4: Road traffic fatalities among car drivers aged 17–24 by road type
source https://publications.parliament.uk/pa/c ... /16905.htm

...DOES show a sudden drop in URBAN fatalities at the right time although RURAL casualties were clearly on a steady downward trend already. That could possibly be down to the pool of vehicles in use, and improving safety standards - I think that's been suggested to be a major factor in the steady reduction of KSIs behind the wheel of cars. I know that the use of older, less-safe, cars by the youngest drivers is a major concern in NZ.

Interestingly, following the link Malc provided, I managed to drill down through several other pages till I got to this one:


Cohort II: A study of learner and new drivers
Publisher: Department for Transport
Publication type: Research report
Published date: 7 May 2008
Mode/topic: Roads, Road safety
This is a major study of learner and new drivers in which ‘cohorts’ of candidates taking the practical driving test were invited to participate in a series of postal surveys. The study provides a picture of how people learn to drive and their attitudes and experiences as new drivers. It also explores the impact of introducing hazard perception testing.

The full dataset is available for research and teaching purposes at the UK Data Archive at:
www.data-archive.ac.uk/findingData/snDe ... sp?sn=5985.

Archived content
The following content is available from the Department for Transport web archive .

Cohort II: A study of learner and new drivers Volume 1 – Main report (PDF – 1,100 KB)
Road Safety Research Report No. 81: Findings Cohort II – a study of learner and new drivers (PDF – 88 KB)
Cohort II: A study of learner and new drivers Volume 2 – questionnaires and data tables (PDF – 4,096 KB)

None of the links to the datasets or PDFs worked.