When we score a course’s Method and Welfare, the single largest factor is whether it builds behavior with reward rather than suppressing it with discomfort. That is not a moral preference dressed up as science. It is where the controlled evidence points. This page lays out that evidence in one place, so our reviews can reference it instead of re-arguing it every time. For the wider picture of how dogs learn, start with the science of dog training.

What “positive reinforcement” actually means

Positive reinforcement is one of the four quadrants of operant conditioning: you add something the dog wants right after a behavior, and the behavior happens more often. The “positive” is not a value judgment, it means adding something, the way a plus sign does. Add food, play, access, or praise that the dog actually values, and you strengthen whatever the dog just did.

The reason it is powerful is timing and clarity. Paired with a marker (a click or a crisp “yes”), reward tells the dog the precise instant it got something right. The dog is not guessing what worked. That clarity is why well-run reward training is fast, and why the mechanical skill of marking and reinforcing is one of the clearest signals that a course is teaching you well.

A common misunderstanding is that reward-based training means permissiveness, endless treats, or ignoring bad behavior. It does not. Good reward training sets clear criteria, uses consequences like withholding the reward or ending the game, and fades food as a behavior becomes reliable. The discipline is real. It is just built on teaching the dog what to do rather than punishing what not to do.

The effectiveness evidence

The honest place to start is that aversive methods can change behavior. Avoiding discomfort is a strong motivator, and suppression can look fast. So the question is not whether corrections work. It is whether they work better than reward, and they do not.

A review of seventeen studies on training methods found no evidence that positive punishment is more effective than positive reinforcement, with some evidence pointing the other way (Ziv, 2017). For the single case people argue about most, off-leash recall, a controlled trial compared dogs trained with electronic collars against dogs trained by reward-focused professionals and found no evidence that the e-collars produced better outcomes (China, Mills and Cooper, 2020). When a tool that causes discomfort does not beat a tool that does not, the burden of proof sits with the discomfort.

The welfare evidence

Here the picture is clearer still, and it is what tips the balance.

A study of ninety-two pet dogs drawn from reward-based, mixed, and aversive-based training schools found that dogs trained with aversive methods showed more stress-related behaviors during training, spent more time in tense and low behavioral states, panted more, and had higher cortisol levels afterward. The effects were not confined to the training session. They showed up in a separate, neutral context later, which suggests the stress followed the dog out of the classroom (Vieira de Castro et al., 2020).

This is why the American Veterinary Society of Animal Behavior’s 2021 position statement recommends reward-based methods as the first line for all training, including for behavior problems, and advises that aversive tools like choke, prong, and electronic collars should not be a first or early choice. The same statement notes a benefit owners tend to care about most: reward-based training is better for the relationship between you and your dog.

Being fair to the other side

We do not think balanced trainers are villains, and our reviews will never treat them that way. Many are highly skilled, many love dogs deeply, and some teach the mechanical craft of training, timing, criteria, clarity, better than their force-free peers do. A few specific points are worth conceding plainly.

A correction can produce a fast, visible result, which is genuinely reinforcing for the owner, not just the dog. Some balanced trainers use aversives at low intensity and with careful conditioning, which is a meaningful difference from a hard leash pop. And there are difficult cases where reward-based progress is slow and frustrating, and a family is out of patience.

None of that changes the weight of the evidence. It changes the tone we owe people. When we review a course built on corrections or an e-collar, we explain why it can feel effective, we lay out the welfare costs the research has documented, and we point to a reward-based approach that reaches the same goal with less risk. We argue with the method, not the person. We go deeper on the most contested tool in our guide to what the evidence says about e-collars.

”But reward training is slower”

This is the most common objection, and it deserves a straight answer. Reward-based training can feel slower at the start, because you are building a behavior up rather than shutting one down. Suppression looks immediate. Construction takes a few sessions.

But “slower to look tidy” is not the same as “slower to work.” Suppressed behavior often returns when the threat of correction is absent, because the dog learned to avoid a consequence, not what to do instead. Reward-trained behavior tends to be more durable because the dog has an actual answer to “what pays here.” And the time you appear to save up front with corrections can be spent later on the fallout: avoidance, anxiety, or a damaged relationship. Measured over the life of the dog, the reward-based route is usually the efficient one, not the indulgent one.

What good reward-based training looks like in a course

When we score Method and Welfare highly, we are looking for specific, checkable things, not vibes:

  • Marker mechanics taught explicitly. The course teaches you to mark the instant of the correct behavior and why timing matters, not just “give a treat.”
  • Clear criteria and progressions. Each skill has a defined goal and a sensible way to raise difficulty, rather than vague encouragement.
  • A plan to fade food. Reward training that never reduces lures or treats was taught incompletely. Good courses show the off-ramp.
  • Emotion taken seriously. The course treats fear and arousal as things to change with counterconditioning, not behaviors to punish.
  • Honest scope. It tells you what reward training does well and where you need more than a video.

How this shows up in our scores

A course does not earn a high Method and Welfare score for saying the words “positive reinforcement.” It earns it by teaching reward mechanics well, respecting the dog’s emotional life, and being honest about limits. A course can also score well on method and still disappoint on teaching, which is why we grade a second axis entirely, covered in why most online courses fail. The full rubric is on our methodology page.

Selected sources: Ziv, G. (2017), Journal of Veterinary Behavior. Vieira de Castro, A.C. et al. (2020), PLOS ONE. China, L., Mills, D.S., Cooper, J.J. (2020), Frontiers in Veterinary Science. American Veterinary Society of Animal Behavior, Position Statement on Humane Dog Training (2021).