If Clicker Training is defined as always giving a reward, then that is not what I do.
That's not what clicker training is defined as.
Clicker training is using a click to MARK a behavior, say, if you were free shaping a roll over, every time the dog moves incrementally to his side, instead of in a sphinx position, you click AND REWARD. Sometimes, if the dog gets confused, you drop criteria until the dog is back to where he was. Eventually (I actually did this in one night with my dog Ozzy) you have a dog that fully understands the concept of roll over.
THEN and ONLY THEN do you add a cue. And you cue the dog every time right before he rolls over. And you reward after every roll over. THEN you start adding variable reinforcements. Sometimes you don't reward after he rolls over. Say, maybe you want him to roll over twice. You hold out on the treat and the click at the end of the behavior until he is doing exactly what you want, when you want, on cue, for no treat, 75 percent of the time.
Then, I consider the behavior trained, but not distraction proofed.
And that's another topic.
Now, I may be wrong, and I may not have explained myself very well, but that's what I do.