So... this is one of my pet peeves about developmental psychology. This "punishment doesn't work" mantra is rampant in the field, and it's totally unsupported

So... this is one of my pet peeves about developmental psychology. This "punishment doesn't work" mantra is rampant in the field, and it's totally unsupported by the scientific evidence. What is true is that punishment on its own is relatively ineffective, and that you need to reward desired behavior to shift it from something undesirable to something desirable. But if you read the learning literature, it's pretty clear that the most effective way to change behavior is to punish undesired behavior and reward desired behavior. The article is promulgating this ideological distortion of operant conditioning that's common in developmental psychology. Maybe there are better or worse ways of implementing punishment, and the best thing is to combine it with rewards; in fact, sometimes rewards on its own will work. But the idea that punishment is ineffective is incorrect.