Breed histories trace the Dane to boar hunting in the Germanic states: dogs built for mass and steady nerve, willing to engage an animal that could seriously injure or kill them. The standard narrative is almost certainly neater than the record allows; what holds from generation to generation is that temperament stayed central as the breed’s role narrowed from hunting to guardianship to the companion dog of today. Breeders pursued that goal with observation, judgment, and a long institutional memory for what a sound animal ought to be — not with behavioral questionnaires or standardized tests, because no such instruments existed for them to use.
As AKC judge Carolyn Thomas has said, "We have very deliberately bred away from dogs with unreliable temperaments." That standard is older than any validated tool on the market. The breed still carries the confidence and presence those programs conserved; what changed is the life we ask these dogs to lead.
The question I keep coming back to is not whether those breeders were naive — they were skilled observers making serious decisions with the means they had — but how much further the same care could go if temperament were as easy to document, compare, and discuss across programs as conformation or health results are today. What would better shared tools actually add?
I can put my hands on a stud dog and tell you whether his structure fits my dams. We have breed standards for that, standardized terminology, and generations of documented pedigree data. I know what a fault looks like and I would never deliberately double down on one. For temperament, the tools are far less developed. Many working breeds are steeped in structured work tests — protection sport, field trials, herding — where rules and repetition surface trainability, nerve under pressure, and how a dog recovers from stress in ways a casual visit rarely does. Training and handler skill still shape what you see, and each sport rewards a different bundle of traits, but the culture puts comparable proof of function on the table in a way informal breeding chats don't. Great Danes were never built around one equivalent, breed-defining job trial; we don't inherit that same depth of standardized functional tradition.
Puppy buyers ask about temperament constantly — will my dog be good with kids, can I raise one in an apartment, will this dog fit my life — and these are fair questions that deserve honest answers. When I went looking for scientific tools I could actually use in my own program, what I found was illuminating but also frustrating.
What Exists
C-BARQ (see Hsu & Serpell, 2003) is the closest thing we have to a gold standard, using 100 validated questions to evaluate 14 behavioral factors plus 22 additional measures. A shorter version, C-BARQ(S), uses 42 questions and was developed for shelter environments — Wilkins et al. validated it using Dog Aging Project participants. Both are scientifically rigorous, but both focus on behavioral problems rather than personality traits, and neither was designed for breeding decisions.
Other questionnaires like the Monash Dog Owner Relationship Scale (MDORS) and Dog Personality Questionnaire ask owners to interpret their dogs' behavior without defining the terms they're using. What one owner calls "aggressive," another calls "alert," and the data that produces is close to useless when you need the kind of precision that breeding decisions demand.
Performance titles demonstrate that a dog can execute specific skills under controlled conditions, but they don't tell you much about how that dog behaves at home or whether its temperament is something you'd want to reproduce.
Hands-on tests like UKC SPOT and ATTS offer standardized evaluations in simulated real-world situations and are probably the most practical tools available to breeders. They're still limited by testing environment and handler influence, though — a dog might pass beautifully on test day and fall apart somewhere unfamiliar with a different person on the lead.
Why None of This Works for Breeding
Most of these tools were built to diagnose problems, and they're good at that. C-BARQ can flag separation anxiety, aggression, and fearfulness with real precision — that's what it was built to measure. That matters in a clinic. But it doesn't help me identify the positive traits I want to breed for — the resilience, the confidence, the willingness to engage with people that makes a dog not just tolerable but genuinely wonderful to live with. A tool built to find what's wrong tells you very little about what's right.
Every tool also measures a snapshot — one day, one environment, one handler. Breeding decisions need something fundamentally different: an understanding of which traits are heritable and how they might express across different contexts and generations. A temperament test tells me how a dog behaves right now. It tells me almost nothing about what that dog will produce.
And none of them account for fit, which is ultimately what determines whether a placement succeeds. A dog that scores high for excitability might be exactly wrong for one family and exactly right for another. The tools measure traits in isolation without helping match a dog to a home, and that matching is half the job.
It's worth noting that the same limitations apply to health testing. Most health tests are snapshots in time, not definitive genetic predictions, and an injury can affect a result without representing anything heritable. The assessment problem runs through everything we do as breeders, not just temperament.
What I've Learned From the Research Anyway
Despite these frustrations, the research does offer insights I've found genuinely useful in thinking about my own program.
C-BARQ's factor analysis shows that behavioral traits cluster in predictable ways — dogs high in stranger-directed fear often show nonsocial fear as well. This means temperament is better understood as interconnected systems rather than a checklist of independent traits, and it has changed how I think about what I'm selecting for.
MDORS makes a point that experienced breeders already know intuitively but that is worth stating plainly: a "good" temperament is really about fit. A high-energy dog isn't inherently a problem dog. A high-energy dog in a sedentary home is a problem dog. The same trait that makes one placement difficult can make another one exceptional.
What C-BARQ contributes, in my view, is specificity. It asks questions like "How does your dog respond when approached by an unfamiliar child while on-leash?" That is a real question with a real, observable answer. Compare that to "Is your dog good with kids?" which means something entirely different to every person who hears it. If breeders are going to get better at communicating about temperament, we need to learn to talk in concrete situations rather than vague labels.
What I Think We Should Be Selecting For
Most research tools focus on problems to avoid. I'm more interested in what to look for, and here is what I think matters:
- Resilience: how quickly a dog bounces back from a bad experience
- Biddability: genuine interest in working with people, not just compliance under pressure
- Stable reactivity: proportional responses to stimuli, neither flat nor explosive
- Social interest that is friendly but controlled
- Environmental confidence: curiosity rather than fear in new situations
- Impulse control and the ability to settle when nothing is happening
- Handling tolerance: accepting grooming, vet visits, and nail trims without a fight
- Size-appropriate behavior, which for Great Danes is not optional
There is no universal perfect temperament, only temperaments that fit well with their intended roles and families. But I believe the traits listed above give a dog the best chance of succeeding in the widest range of homes.
The Great Dane Problem
A fearful 20-pound dog hides under furniture. A fearful 140-pound Great Dane panics and someone gets hurt. Our breed's size turns temperament issues that might be manageable in a smaller dog into genuine safety concerns, and this should shape how seriously we take temperament in our breeding decisions.
The 2004 GDCA National Health Survey, covering 519 households and 1,564 Great Danes, puts numbers to this gap. Twenty-seven percent of dogs in the survey had at least one reported temperament issue, including 12% with dog-directed aggression, 8% fearfulness, and 5% human-directed aggression. But only 11% of households performed any formal temperament testing. When owners did use screening results to eliminate dogs from breeding, temperament ranked 4th or 5th on the list, well behind hip X-rays, cardiac ultrasound, and thyroid screening (Slater, 2004).
That gap between the prevalence of temperament issues and the frequency of temperament screening is the whole problem captured in one data set. We screen for hips. We screen for hearts. We barely screen for the thing puppy buyers ask about most, and the thing that determines more than anything else whether a placement will succeed or fail.
Just as we would never breed two dogs with the same conformational weakness, we should not be unknowingly concentrating negative temperament traits. But without better tools and a shared language for discussing what we observe, we risk doing exactly that. Imagine if we could talk about temperament with the same precision we use for OFA results — if we could share standardized behavioral information the way we share hip scores and cardiac clearances. We are not there yet, but that is the direction worth moving in.
Scientists are collecting data through projects like Darwin's Ark and the Dog Aging Project, and the Functional Breeding Podcast has discussed how difficult it remains to translate any of this into practical breeding decisions. They're right — it is extremely difficult. The practical takeaway for breeders right now is simpler: ask specific questions about specific situations, observe your dogs across varied contexts, and follow up with your puppy buyers over the long term. What you learn from a two-year-old dog living in a family home is worth more than any test score taken on a single afternoon.
We have always done this work with limited tools. The breed's own history proves that temperament selection works — breeders took a boar hunter and produced a gentle giant without a single validated questionnaire. The hope is that better tools are coming. In the meantime, we keep watching, keep asking, and keep learning from the dogs and the families that raise them.
Sources
Primary Research:
Hsu, Y., & Serpell, J. A. (2003). Development and validation of a questionnaire for measuring behavior and temperament traits in pet dogs. Journal of the American Veterinary Medical Association, 223(9), 1293-1300. DOI: 10.2460/javma.2003.223.1293
Wilkins, V., Evans, J., Park, C., The Dog Aging Project Consortium, Fitzpatrick, A. L., Creevy, K. E., & Ruple, A. (2024). Validation of the shortened version of the Canine Behavioral Assessment and Research Questionnaire (C-BARQ) using participants from the Dog Aging Project. PLoS ONE, 19(4), e0299973. DOI: 10.1371/journal.pone.0299973
Dwyer, F., Bennett, P. C., & Coleman, G. J. (2006). Development of the Monash Dog Owner Relationship Scale (MDORS). Anthrozoös, 19(3), 243-256. DOI: 10.2752/089279306785415592
Slater, M. R. (2004). Great Dane Club of America National Health Survey. Texas A&M University College of Veterinary Medicine. Conducted for the GDCA Health and Welfare Committee. Survey data collected 2001–2003; final report January 19, 2004.
Additional Discussion:
Functional Breeding Podcast. (2025, March 4). "Hekman and Stremming: New paper on prevalence of behavior problems in dogs in the US." Podcast episode
Thomas, C. Quoted in "Great Dane History: The Apollo of Dogs." American Kennel Club.
Resources:
- C-BARQ: University of Pennsylvania
- Darwin's Ark: darwinsark.org
- UKC SPOT Program: ukcdogs.com
- ATTS Temperament Test: atts.org