Some thoughts on correspondence between saying and doing.

Israel (1978) looked at early studies on correspondence training. That is when a child says, “I will put the toy away” and then actually does it.

The paper asked: Do we really know what we are measuring? It found fuzzy terms, weak methods, and little proof that the training works.

The author found big holes. Studies used different words for the same thing. Some counted only talk. Some counted only action. Few checked if talk matched action later.

Because of this, the field could not tell if the training caused real change. C warned: fix the science first, then use it in schools and clinics.

Malott (2018) answered the warning. That paper built a trainer model that puts JABA and JEAB articles ahead of cook-book protocols. It shows how to grow science-first staff.

Eugenia Gras et al. (2003) and Najdowski et al. (2003) pushed the same idea into policy. They told psychology to stop selling brand-name therapy lists and start teaching basic principles instead.

Geckeler et al. (2000) tracked the field from 1990-1999. They saw verbal-behavior studies start to cite fresh data, not just Skinner 1957. This shift supports C’s call for newer, cleaner evidence.

If you run correspondence programs, pause and check your definitions. Write the target words and the target acts in plain, matching language. Then measure both every time. This small step moves your case from hopeful story to solid science.