E COLLARS!
- RuffStartDogTraining
- Aug 6
- 7 min read
Updated: Aug 8
That should get the algorithm going.
I've Read the Science. Because I Want You To.
Now let's get something out of the way before anyone rushes to the comments.
I don't use e-collars.
Where I live in Wales, their use on dogs is prohibited.
But even if they weren't, I still wouldn't use one.
Partly because of my own views on dog training.
And partly because I don't believe I have the experience to use one safely.
I actually think that's something more trainers should be comfortable saying.
If I don't have the knowledge or experience to use a tool properly, I don't use it.
Simple.
That doesn't make me morally superior.
It doesn't mean everyone who uses an e-collar is cruel.
And it certainly doesn't mean I know everything.
It simply means I've made a professional decision about the tools I'm prepared to use.
What follows is merely an academic discussion where I take a closer look at the evidence.
Recently I've noticed the e-collar debate flaring up again.
As always, the same claims started appearing.
"The science proves they're cruel."
"The science proves they work."
"They're just a communication tool."
"They're abuse."
"Positive reinforcement can't solve everything."
"Anyone who uses one shouldn't be allowed near a dog."
(Just a quick edit. There are people who abuse dogs without any tools there are people who abuse dogs with tools. Abuse is abuse. This isnt about that.)
After a while it all starts sounding depressingly familiar.
Everyone seems absolutely certain.
So I decided to go back to something I haven't done for a while.
I started reading the research.
Not summaries.
Not Facebook memes.
Not YouTube videos.
The actual papers.
And something surprised me.
The biggest thing that struck me
If you'd asked me before I started reading the studies what I expected to find, I'd probably have said I expected one side of the argument to be clearly supported by the evidence.
That's not what I found.
The biggest thing that struck me wasn't whether e-collars work.
Or whether they don't.
It was just how difficult behavioural science really is.
Good science should be repeatable.
But behavioural science has a problem that my years working with chemicals didn't.
Dogs aren't test tubes.
Every dog is different.
Every owner is different.
Every trainer is different.
Every environment is different.
Every learning history is different.
Breed matters.
Genetics matter.
Previous experiences matter.
Stress matters.
Motivation matters.
Even the weather can matter.
When I started comparing the studies, I quickly realised something.
They weren't repeating the same experiment.
They weren't even asking the same question.
Some looked at welfare.
Some looked at recall.
Some looked at chasing behaviour.
Some measured physiological stress.
Some asked owners whether they thought the training had worked.
Others objectively measured performance.
Some involved professional trainers.
Others looked at ordinary dog owners.
Some involved dogs already referred for behavioural problems.
Others selected dogs because they naturally wanted to chase.
Those aren't identical experiments.
They're pieces of a much bigger puzzle.
And I think that's where both sides of this debate sometimes go wrong.
One study becomes "proof."
Another gets ignored because it doesn't support the argument.
But science doesn't work like that.
The studies at a glance |
Study | Participants | who used the e-collar? | Main finding | |||
Blackwell et al. (2012) | Nearly 4,000 owner surveys | Owners | Reward-based owners reported greater success than owners using remote e-collars, but this was an observational survey rather than a controlled trial. | |||
Cooper et al. (2014) | 63 dogs | Professional trainers | No clear evidence that e-collars produced better overall outcomes than the comparison groups. Welfare concerns were also identified. | |||
China, Mills & Cooper (2020) | 63 dogs | Professional trainers | No overall training advantage for the e-collar group. Some measures favoured the reward-based group. The paper has since attracted published methodological criticism. | |||
Johnson & Wynne (2024) | 30 dogs recruited; 19 completed | Researchers/trainers | The e-collar group was more effective at stopping a specific chasing behaviour under experimental conditions, but the final sample was small and highly selected. |
Straight away, three thing stand out.
First These studies aren't all investigating the same thing.
They're asking different questions.
If they're asking different questions, we should be careful about pretending they all give the same answer.
Second The sample size. there are approximately 13 million dogs living in the UK.
That's like asking what men want but only asking 15 year old boys.
Third they cannot be replicated.
What the studies actually tell us
The 2014 DEFRA-funded study is probably the one most people have heard about.
It looked at 63 pet dogs that had already been referred because of behavioural problems such as poor recall, chasing livestock or aggression towards other dogs.
The important detail is this:
The e-collars weren't handed to ordinary owners.
Professional trainers used them.
The study found no clear overall advantage for e-collar training over the alternatives and also identified behavioural indicators that raised welfare concerns.
That's important.
But it's equally important to recognise what it doesn't tell us.
It doesn't tell us what every trainer would achieve.
It doesn't tell us what every dog would do.
And it certainly doesn't tell us what happens when someone buys an e-collar online and starts using it themselves.
The 2020 paper analysed data from the same wider research programme.
Again, the dogs were trained by professionals.
Again, no clear overall advantage was found for the e-collar group.
The reward-based trainers actually performed better on some of the measured outcomes.
But this paper has also been criticised by other researchers, who questioned aspects of the methodology and interpretation.
That's not unusual.
That's science.
Researchers are supposed to challenge each other's work.
Then there's the 2024 study.
This is the one that many supporters of e-collars now point to.
And I think they're right to point to it.
Because if we're going to be honest about the science, we can't ignore findings simply because they make us uncomfortable.
The study found that, under very specific experimental conditions, the e-collar group stopped dogs chasing a moving lure more effectively than the food-reward groups.
That's a genuine finding.
But context matters.
The study began with 30 dogs.
Only 19 completed the chasing experiment.
The final e-collar group contained just six dogs.
Those dogs had also been specifically selected because they naturally wanted to chase.
So does the study show that e-collars can be effective under carefully controlled conditions for a very specific behaviour?
Yes.
Does it prove they're the best answer for every behavioural problem?
No.
Does it prove ordinary owners can achieve the same results?
Again, no.
Finally, there was the 2012 Blackwell survey.
Nearly 4,000 owners responded.
Among them, 133 reported using remote e-collars.
The reward-based owners reported greater success than those using e-collars.
That's interesting.
But it wasn't a controlled experiment.
The owners chose their own training methods.
The researchers weren't watching them.
The results relied on owner reports.
So we have to be careful about claiming cause and effect.
Here's what I think both sides are getting wrong
The people who oppose e-collars often present the research as though it proves they're ineffective, cruel and unnecessary in every circumstance.
The people who support them often present the research as though it proves they're safe, highly effective and misunderstood.
I don't think either position is supported by the evidence.
The research is much more nuanced than that.
Some studies raise genuine welfare concerns.
Some fail to show a clear advantage over reward-based training.
One recent study demonstrates impressive results for a very specific behaviour under tightly controlled conditions.
None of those findings automatically cancels out the others.
And none of them allows us to make sweeping statements like:
"The science has settled this."
I'm not convinced it has.
What I think the science has done is show us just how complicated behaviour really is.
So where does that leave me?
Exactly where I started.
I don't use e-collars.
I wouldn't use them even if they were legal here.
Not because I think every person who uses one is cruel.
Not because I think the science is black and white.
But because I know my own limitations.
I don't have the knowledge or practical experience to use one safely.
And if I'm honest, I don't think enough trainers say that.
One thing my background in science taught me is that certainty can be dangerous.
Good scientists don't usually say:
"I've proved I'm right."
They say:
"Here's the evidence. Here are its strengths. Here are its limitations. And here's what we still don't know."
I think the dog training world could benefit from a little more of that mindset.
So don't take my word for it.
Read the studies.
Read the criticisms.
Read the papers that agree with you.
Then read the ones that don't.
Because the more I read, the less interested I am in winning an argument.
I'm far more interested in understanding what the science was actually saying.
I have a feeling that's something we could all do a little more of.
Science isn't a courtroom. It isn't about proving one side guilty and the other innocent. It's a process of reducing uncertainty. And after reading these papers, I don't feel more certain—I feel better informed.
If this article encourages just one person to read the research before repeating, "The science says...", then it's done exactly what I hoped it would.
If I've misunderstood any of these papers, or if you've found research I've missed, I'd genuinely like to read it. Science moves forward through discussion, not by pretending we've already got all the answers.
Further Reading
If you're interested in reading the papers yourself rather than relying on social media summaries, here are the main studies discussed in this article.
Blackwell et al. (2012) – The use of electronic collars for training domestic dogs: estimated prevalence, reasons and risk factors for use, and owner perceived success as compared to other training methods
Cooper et al. (2014) – The Welfare Consequences and Efficacy of Training Pet Dogs with Remote Electronic Training Collars in Comparison to Reward Based Training
China, Mills & Cooper (2020) – Efficacy of Dog Training With and Without Remote Electronic Collars vs. a Focus on Positive Reinforcement
Johnson & Wynne (2024) – Comparison of the Efficacy and Welfare of Different Training Methods in Stopping Chasing Behavior in Dogs
Sargisson & McLean (2021) – Published commentary discussing aspects of the 2020 paper's methodology and interpretation.
Comments