A procedural analysis of correspondence training techniques.
Name every detail of your correspondence-training procedure—who talks, who checks, when reinforcement lands—or your “replication” will be a different program in disguise.
01Research in Context
What this study did
The author took every correspondence-training paper he could find. He pulled apart who said what, who checked, and when reinforcement hit.
He built a simple grid with four questions: Who reports? When do they report? Who checks? When is reinforcement given? Every study now fits one box per question.
No new kids, no new data—just sharper labels so future studies can line up like silverware instead of tossed socks.
What they found
Old labels like “say-do” and “show-do” hide the real moving parts. The same nickname can cover opposite procedures.
The new grid shows the hidden mix-ups. Once you name each dimension, you see why some studies replicate and others flop.
How this fits with other research
Reiss et al. (1982) used “say-do” to fix posture in a classroom. Under the new grid their procedure is: child reports, adult checks, reinforcement after activity. A later team copying “say-do” could swap any piece and think they failed to replicate—when they really ran a different procedure.
Lord et al. (1986) kept kids on track after training by thinning reinforcement and keeping the child’s rule statements on cue. The grid labels this “post-training reinforcement plus child report,” a combo the taxonomy now lets you name and test separately.
Lloyd (2002) complained that correspondence research “stalled” after 1992 because procedures were messy. The 1990 paper hands that review the exact labeling tool it asked for, so the stall can end.
Why it matters
Next time you set up a correspondence program, fill in the four boxes first. Write who reports, when, who checks, and when you reinforce. Share that grid with colleagues and future replicators. Clear labels keep your good results from vanishing when the next therapist tries to “do the same thing.”
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open your last correspondence plan and tag each step with the four grid questions—then share the filled grid with your team before the next session.
02At a glance
03Original abstract
A variety of names have been given to procedures used in correspondence training, some more descriptive than others. In this article I argue that a terminology more accurately describing actual procedures, rather than the conceptual function that those procedures are assumed to serve, would benefit the area of correspondence training. I identify two documented procedures during the reinforcement of verbalization phase and five procedures during the reinforcement of correspondence phase and suggest that those procedures can be classified, or grouped into nonoverlapping categories, by specifying the critical dimensions of those procedures belonging to a single category. I suggest that the names of such nonoverlapping categories should clearly specify the dimensions on which the classification is based in order to facilitate experimental comparison of procedures, and to be able to recognize when a new procedure (as opposed to a variant of one already in existence) is developed. Future research involving comparative analysis across and within procedures is discussed within the framework of the proposed classification.
The Behavior analyst, 1990 · doi:10.1007/BF03392528