21/04/2026
DOG TRAINING MYTHS that I wish would die ...
I've just been writing a post analysing a particularly insightful study from Masson, Nigron & Gaultier (2018).
This is one study I come back to over and over again because it's such a rich piece of work.
They asked 1251 people for the training methods they used with their dog. No judgement. No implied 'perfect' answer. 330 people said they used some form of shock collar on their dog. Some used remote control shock collars. Others used bark-activated collars. Some used invisible perimeter fences that zapped the dog if they went near a buried activation wire that emitted a signal that activated the collar.
One of the big questions they asked was whether the type of shock collar worked.
And they went out of their way to ask in a neutral way that implied no correct answer so people were more likely to give their honest opinions if they'd used the collar.
How successful the collar was depended on the type of collar, as you might imagine. The least effective were bark-activated collars. They were only successful 25% of the time.
Even the most "successful" collars - the remote-activated ones - were, at best, only effective 51% of the time with certain types of problem.
No offence, but if I marketed a product that was that s**t, who'd buy it?
Say you need your car to get to work, but it only starts half the time you need it.
Or you go for a bus that only turns up half the time.
Or you buy a packet of crisps, and half the time it's filled with nothing but air.
And that's at best.
At worst, you need your car but it only starts once every four times. And that bus? It only turns up one day out of four. That packet of crisps? Three-quarters of the time, you've paid for nothing but air.
Masson and colleagues didn't just ask whether it worked, they also asked whether the collar injured the dog. The worst collars for injuries were bark-activated collars.
12% of the time, they caused some sort of burn-related injury.
So you're not just trying to turn your car on and it fails three days out of four, it also gives you caustic burns or some such every couple of weeks.
That's not really the astonishing bit for me.
The bit that always gets me is the fact that loads of people THOUGHT collars could be effective. Of the 330 people who said they had used some form of electric collar, much higher percentages of people THOUGHT they worked (in general) to solve problematic behaviours in dogs than had actually had success with them.
That's one hell of a bias.
It fails you, so you think it will work for other people?!
And that was also true of the people who hadn't used one. Even though they'd had success with the approaches they used, they THOUGHT they were more effective than people said they actually were.
It's not just you with a car that fails to start three mornings out of four. It's all your neighbours. But you ALL think that cars in general are good at getting you from A to B.
That's quite mad.
When advocates of shock collars talk, you'd think they're giving 100% success rates. I've heard two "trainers" talking in the UK about how shock collars give dogs liberty and "re-train" them to reduce predation as well as improving recall.
Yes, the, 'Won't someone PLEASE think of the sheep?!' pearl clutchers.
Can you imagine if their views had to be accompanied by clearly stated figures that 49% of the time or for 49% of dogs, the collar will fail when you most need it to work?
And you know which dogs it would likely work with?
The ones whose recall was occasionally so-so.
The ones who bothered livestock and wildlife.
Dogs like my girl Tilly, admittedly a bit of a chicken chaser on her arrival with me.
She was quite persistent. Cockers are going to cocker, after all.
Within about five minutes of on-lead stuff over about five days, Tilly had quickly realised that hens are not for chasing. Those hens soon learned that Tilly was a slightly annoying occasional egg thief and competitor for corn.
I can't help but think that she'd be the kind of easy dog who would be quite easily corrected by a collar (and then develop a bunch of problematic behaviours as a consequence).
If it had failed with her, I'm sure the worst that would have happened was some frantic blonde chasing, a lot of hoo-hah with the hens, perhaps even a lot of pecking.
I absolutely would not want to test a shock collar on a dog where failure would be catastrophic. Like my dog Lidy. If a collar failed with her... well I already know what the outcome would be. It's a hideous outcome on all sides. Thankfully a lead, a muzzle and a bit of training are more than enough to prevent all emergencies, like the three sheep that had got out of the field this morning.
I wouldn't want to put my faith in something that's likely to fail.
So that's one training myth I really wish would die. I really wish people would stop thinking that shock collars are efficient. That's not what surveys say.
Masson and colleagues conclude similarly. They said that communication around these devices needed to be clearer. Since 75% people bought it without supervision or guidance, buying from the internet or from a pet store, one recommendation that could be drawn was that sales needed to be better regulated and purchasers should be aware of two things: the propensity to cause burns and the likelihood of failure.
Many countries in Europe have thankfully gone much further than regulation of sales since this study came out. I'm glad about that. Hearing people selling snake oil to desperate purchasers always grinds my gears.
The problem is much deeper than that though.
The problem is that we believe they're more effective than they are.
We believe they can solve problems they don't.
And that's one myth I wish would die.
Masson, S., Nigron, I., & Gaultier, E. (2018). Questionnaire survey on the use of different e-collar types in France in everyday life with a view to providing recommendations for possible future regulations. Journal of Veterinary Behavior, 26, 48-60.
