Demystifying moderators and mediators in intellectual and developmental disabilities research: a primer and review of the literature.

Farmer (2012) wrote a how-to guide for researchers who study people with intellectual or developmental disabilities. The paper walks readers through two big statistical tools: moderator analysis and mediator analysis.

Moderators tell you "for whom" or "when" a treatment works. Mediators tell you "why" it works. The author searched the IDD literature and found few studies using either tool.

Most IDD studies still ask only "Does it work?" They rarely ask "Who benefits most?" or "What skills explain the gain?" Without these follow-ups, we can’t refine treatments.

The review shows that wider use of moderator and mediator tests would speed up progress in the field.

Robinson et al. (2011) looked at the same problem one year earlier. They warned that standard drug-style RCTs fit poorly with IDD behavioral work. Farmer (2012) builds on that warning and gives a concrete fix: use moderator and mediator models instead of simple yes-no trials.

Abbeduto (2014) and Oliver (2014) later echoed the same message. Both editorials say the field needs sharper measures and smarter designs. They treat the Farmer (2012) primer as a key stepping-stone toward those goals.

Zwiya et al. (2023) push the idea even further. They ask for participatory, equity-centered research. Their call for justice-minded methods extends the statistical cleanup that Farmer (2012) started.

If you run or read IDD studies, this paper gives you a checklist. Add a moderator test to see if age, IQ, or setting changes the outcome. Add a mediator test to learn whether gains in communication drive drops in problem behavior. These extra steps turn a flat "it worked" into a roadmap for better therapy.