19/03/2026
Thank you Julie
You’ve might have seen today’s headline: “Crossbreed dogs show more behavioural problems than pure breeds, study suggests.” It’s already flying around social media and I can practically hear the “told you so” crowd warming up.
I read the actual study this morning. The headline findings didn’t sit right with me. Something didn’t add up.
So I decided to pore over it properly - the methods, the variables, the statistical models, the references. And the more I read, the more problems I found.
Honestly, I really wanted this study to be good. Everyone who works with dogs knows that online groups and forums are full of people struggling with their cockapoos.
But then, every daycare, every dog walker, every dog-friendly pub in the UK is also full of cockapoos. Because they’re incredibly popular.
The question has always been, are these dogs genuinely more problematic, or are we just seeing more of them everywhere, including in the “problem” pile? That’s a popularity versus prevalence question, and it’s a really important one.
So when I saw a study that appeared to be trying to separate those two things out, I was genuinely hopeful. Finally, some proper data.
It didn’t deliver.
Let me tell you why I’m someone who would notice that.
I’ve spent over 15 years as a specialist in canine separation anxiety. I’ve seen tens of thousands of dogs in my community. And a huge number of those have been cockapoos, cavapoos, and labradoodles.
The question of whether these dogs have more behavioural problems than purebreds comes up constantly in my work. It’s a question I’ve been actively interested in for years. I’ve read the research on this topic extensively.
I shared my life for 15 years with two amazing cockapoos.
So I have both the professional and personal interest to take this study seriously.
And I did take it seriously. Which is exactly why I have problems with it.
To be fair, there are things to credit here. This is a large study - over 9,000 dogs- and they used a validated, standardised behaviour questionnaire called the C-BARQ.
They collected a lot of useful data on training methods, owner demographics, and breeding background. The ambition is genuinely welcome. We need more research like this, we really do.
And here's what they concluded. (Cross breed dog parents brace yourselves...)
They found that cockapoos and cavapoos scored worse on several behaviour measures - things like excitability, fear, separation-related problems, and aggression.
Labradoodles were more of a mixed bag, actually scoring better than poodles on several measures.
Sounds straightforward, right? Crossbreeds behave worse. Case closed.
Except it’s not. Because here’s what the study didn’t do, and this is where it falls apart.
**They ignored what happens when a breed gets popular. And it’s the key to this whole study.**
This is a big one. Every dog in this study was acquired from 2019 onwards. That means the entire crossbreed sample was purchased during or after the COVID puppy boom.
This is the exact the period when demand for cockapoos, cavapoos, and labradoodles massively outstripped the supply of responsibly bred puppies.
We all know what happens when a breed explodes in popularity. You get a flood of inexperienced, profit-driven breeders producing puppies with no screening for temperament, no health testing, no proper early socialisation, and no support for buyers.
The market gets swamped with dogs from backyard breeders and puppy farms. This isn’t speculation...it’s exactly what happened with these crossbreeds during COVID.
And here’s why that matters so much. Those of us who work in dog behaviour have long known that where a dog comes from matters enormously.
We have solid research on this such as McMillan’s work on pet store dogs, Wauthier and Williams on puppy farming, among other.
And this solid research shows that dogs from poor welfare breeding operations (puppy farming, backyard breeders and overseas puppy mills) show significantly higher rates of aggression, fear, and separation problems.
This is one of the most well-established findings in canine behavioural research.
So did the researchers measure breeding quality properly? No. They used one question. A single yes-or-no: “Did you see the puppy’s mother when you collected it?”
That’s it.
Anyone who works in dog welfare knows that “meet the mum” stopped being a reliable indicator of a good breeder years ago.
Every puppy farmer and backyard breeder everywhere now knows to have a mother dog on site. It’s the first thing they learned to fake.
Using this as your measure of breeding quality in 2026 is like using “do they have a website?” to identify a legitimate business.
And if you’re reading this thinking “oh no, did I buy from one of these places?” — please don’t beat yourself up.
The reality is that unscrupulous breeders have become incredibly sophisticated. They have beautiful websites, lovely photos, convincing stories, clean-looking homes.
They have learned exactly what to say and show to make you feel confident you’re buying from a good breeder.
They are deliberately designed to be hard to spot. Plenty of experienced dog people have been caught out too.
This isn’t about blaming owners - it’s about an industry that has become expert at deception.
The study collected no data on where the dog was actually purchased, whether it was from a Kennel Club registered breeder, a breed club referral, a Pets4Homes ad, or a Facebook marketplace listing.
No data on whether health tests were done on the parents. No data on whether there was a waiting list. No data on the breeder’s experience. No data on early socialisation. Nothing.
They also included no comparison with other breeds that went through the same popularity explosion.
French bulldogs, for example, experienced an almost identical demand surge, breeding quality collapse, and first-time-owner demographic shift during the same period.
(Side note: my own client list and Facebook group are also full of French bulldogs and dachshunds — other breeds that have surged in popularity. Funny, that.)
If they’d included French bulldogs and found the same elevated behaviour scores, that would have told us the story is about popularity-driven breeding issues, not about crossbreeding.
But they didn’t. Which means they literally cannot tell the difference between “crossbreeds have worse behaviour” and “dogs from the unregulated puppy market have worse behaviour.”
And that’s just the biggest problem. There are more.
1. They didn’t adequately account for owner experience.
The study found that 50% of crossbreed owners were first-time dog owners, compared to 34% of purebred owners. That’s an enormous difference.
First-time owners are consistently shown in research to report more behavioural problems. Partly because they may genuinely struggle more with training, but also because they’re more likely to interpret normal dog behaviour as problematic.
I know this from personal experience. The only reason I recognised that my cockapoo Percy’s separation anxiety wasn’t normal was because I’d already lived through a very different experience with my first dog, India. I had a benchmark.
If Percy had been my first dog, I might not have spotted it — or I might have assumed all dogs were like that. Most first-time cockapoo owners don’t have that comparison point, and nobody should expect them to.
A cockapoo doing zoomies when the lead comes out is not a behavioural problem. It’s a normal dog being excited about a walk.
But if you were told to expect a calm, easy, low-maintenance companion - because that’s what the breeder's photos on IG suggested -- you might well score that as “high excitability” on a questionnaire.
That’s not an owner failing. That’s an owner who was sold a lie by an industry that profits from unrealistic promises.
The researchers tried to adjust for first-time ownership statistically, but they treated it as a simple yes/no variable. There’s a world of difference between different types of first-time owner, and you can’t capture that with a tick box.
2. They didn’t include training as a variable in their behaviour models.
The study collected data on training methods. They know that cocker spaniel and labrador retriever owners were significantly more likely to use a combination of reward and aversive training methods compared to the crossbreed owners, who used more rewards-only training.
And yet they didn’t include training methods in their statistical models. So they’re comparing groups who trained their dogs in fundamentally different ways, and attributing the behavioural differences to breed genetics. That’s a glaring omission.
And by the way, a shout out to all you first-time dog owners who are choosing to train your dogs with kindness. The study found that crossbreed owners were more likely to use reward-based methods than the purebred gun dog owners. That’s something to be proud of, not penalised for in a behaviour comparison.
3. The questionnaire can’t distinguish between suppressed behaviour and genuinely transformed behaviour.
Here’s where it gets really interesting. The study itself acknowledges that better “trainability” scores in the purebred gun dogs may reflect forced suppression of behaviour through aversive methods rather than genuinely better temperament.
They actually wrote that. Suppressed behaviour not improved behaviour.
But then they still treated those scores as evidence that the purebreds were better behaved.
Suppressing behaviour and helping a dog do things differently are not the same thing.
A dog trained with punishment may appear more “obedient” on a questionnaire, but that doesn’t mean they’re better behaved. It means they’ve learned to shut down. That’s not a good outcome. It’s a welfare concern.
And before anyone jumps on this as evidence that aversive training “works”, no. The research on the harmful effects of punishment-based training is extensive and clear. What this study shows is that aversive methods can suppress the expression of behaviour enough to skew a questionnaire. That’s not the same as having a well-adjusted dog. Not even close.
4. Their genetic argument contradicts itself.
The study tries to explain cockapoo aggression scores by pointing to historic research on aggression in cocker spaniels namely the so-called “Cocker Rage” literature.
But that research is weak, outdated, and primarily relates to specific colour lines in English cocker spaniels, not the breed as a whole.
More importantly, their own data shows cocker spaniels scoring well on the aggression measures. So if cocker spaniels don’t have an aggression problem in their own data, how can cocker spaniel genetics explain aggression in cockapoos?
And crucially if you accept basic genetics, crossing two breeds should produce offspring that fall somewhere between the two parents. That’s regression to the mean.
So even if cocker spaniels did carry aggression-related genes, you’d expect cockapoos to be less aggressive than cocker spaniels, not more.
The genetic argument doesn’t just fail to support their conclusion. It actively contradicts it.
So what does this study actually show?
It shows that dogs acquired during a puppy boom, disproportionately from unregulated breeders, by first-time owners with novice expectations and less training experience, score worse on an owner-reported behaviour questionnaire than established purebred breeds with decades of structured breeding programs and more experienced owners.
That's it. Not a thing more.
That’s not a finding about crossbreeds. That’s a finding about the puppy market. And it’s a finding that those of us working in dog behaviour and welfare have been shouting about for years.
If you’re a cockapoo, cavapoo, or labradoodle owner reading this, please hear me...
-Your dog is not defective.
-Your dog is not genetically doomed to be badly behaved.
- And you are not a bad owner.
If your dog has behavioural challenges, that doesn’t mean you’ve failed. It just means you’re dealing with something that may well have started long before your dog came home to you.
Breeding matters. Early life matters. And neither of those things was in your control.
And it could also be that you’re simply an attuned, sensitive dog parent - that you’re genuinely concerned about your dog and invested in their behaviour.
That doesn’t make you or your dog a problem. That makes you a good owner.
What this study should make you angry about isn’t your dog’s breed. It’s an unregulated breeding industry that has been profiting from these dogs while setting them and their families up to struggle.
And if you’re someone about to smugly share that Guardian headline in your purebred breed group please read the actual study first.
*The study referenced is: Bryson et al. (2026), “Comparing undesirable behaviours between ‘designer’ Poodle-cross dogs and their purebred progenitor breeds,” published in PLOS ONE.*
