: How to teach a dog to sit and stay There are obviously many ways to do so, but what is the most efficient and best for the dog ? The idea is to have the dog sit on a verbal and/or visual cue and then stay in position until

I'll base my answer on my recent experience with positive reinforcement training, aka clicker training to teach my puppy a sit and stay.

First, a few words on that:

Positive reinforcement training is based mainly on positive rewards, that means you will condition your dog that what you want him to do is rewarded and has positive consequences. That means when the dog does the behaviour you reward him with treats, affection, play, toys, etc. Treats work best with most dogs and are easy to use in a training session, and you can have a good reinforcement rate with no interruptions.
Use of a marker: positive reinforcement training is often paired with the use of a marker. That marker is the "clicker" part in "clicker training". The mark is conditioned to the reward. To the dog, click means "what I'm doing right now is good, so a reward is coming". This is called classical conditioning (Pavlov conditioning). It has to be done before training any particular behaviour. Usually 2-3 sessions of just "click => treat" without asking anything to the dog is sufficient. The marker conveys information to the dog and so timing is crucial. The marker is also a reward in itself.
This method avoids any use of physical or psychological punishment (technically positive punishment, something bad is added) which has many advantages (outside the scope of this question) and it also doesn't consider lengthy discussions on "dominance" or question why the dog is not doing the behaviour. If the dog doesn't do what you want, it just doesn't get rewarded and you try again.
Studies showed that reward-based training methods lead to better obedience than punishment-only methods (see for example Hiby et al.)

This being said, how do apply that to teaching a sit and stay. What we want to obtain is this: you cue the dog (verbally "sit" or with a hand gesture) and he sits. And stays in that position. You can move around, he just stays. Then you release him with another cue ("free" or "release" for example).

This is different from the usual "sit" where you sit the dog and then ask him to stay with "stay".

To teach that we have to proceed in a few steps:

Get the behaviour
Put the sit under "stimulus control" and teach the release cue
Add duration, distractions and

1. Get the behaviour

This can be done by "capturing" or luring.

Capturing: Take your clicker and treats and wait. When the dog sits, click and treat. Toss the treat away from the dog so that he moves around and then wait again until he sits. Repeat multiple times. Change room, try again.

Luring: Take a treat in your hand and click and treat the dog for following the treat. Then lure the dog into a sit, click and treat. Quickly you have to remove the treat. Pretend to have a treat in your hand and do the same.

At that point the dog will understand the behaviour very clearly and you can practice a combination of both methods over a few sessions. It is important to let the dog think about what he has to do. Luring with the treat just get the behaviour started, then fade the treat. That also provides a good brain exercice session for your dog, similar to humans playing Sudoku ?

For both methods, proceed in small steps: first click and treat for muscle movements (the dog is about to sit), then for his legs moving, then when he's properly sitting. That the key point and that's where your timing when you click is important. It is really teaching the dog how to perform the behaviour you want. For a sit that might seem of no importance, but for more complicated behaviours that's where the "clicker magic" happens.

2. Stimulus control and release cue

Now you will add a cue to the behaviour. My preferred method at this point is to lure the dog (with and then without an actual treat in my hand) into a sit. But just before doing that, give your cue. Practice by saying "sit", then lure into a sit. Or practice by doing your hand gesture and then lure into a sit.

The timing of your click should be as before, start again with small steps. If after a few sessions you want to click only when the dog sits but he's not doing it, go back a step.

When you are at the point where you click for a complete sit, simultaneously always release the dog from the sit. He sits, you click and treat, then give your release cue and let the dog move. You can also say your release cue, toss a treat away from the dog (to make him move) and click.

3. Duration, distractions

Now you have a reliable sit behaviour after your sit cue, but if you move around or wait too long the dog starts moving.

You have to separately and very progressively add duration and distractions.

Duration: Ask for a sit, wait a bit then click and treat. The dog stays in a sit. Wait a little bit more, give your release cue, the dog moves, you click and treat.

Distractions: Proceed step by step, first clicking when the distraction happens, then after the distraction has happened.

Proceeding in that way makes the whole thing a chain of reinforcement, with the last element of the chain being the one with the more reinforcements. You cue the dog to sit, he's likely to do so as he knows reinforcements will follow. Then he waits, as your release cue is another reinforcer, so he waits because that's the way he gets the rewards, then your release him and that end the chain, with being free + primary reinforcer (treat, petting, etc.). That also explains why you can get the whole behaviour (once learned) with no clicker and no treats. Each part of the chain is a reinforcer for the previous one. At the end you a simple "good boy/girl", petting, playing with a toy or simply being free to continue the walk will do the trick

4. Add a bow tie

References

You can find much more on this whole subject in this Youtube video.

The foundations and principles of positive reinforcement training and clicker training are covered at length in the book "Don't shoot the dog" by Karen Pryor. Another one on the principles and applications of clicker training is "Reaching the animal mind" by the same author.

I've also found this video from Michael Ellis. I didn't know him before but he seems to be a very experienced competition dog trainer. His training method is not said to be clicker training, however the concepts are exactly the same. Worth watching.

---