Matching, statistics, and common sense.
The matching law survived early statistical attacks and still guides both lab and real-world choice analyses.
01Research in Context
What this study did
Haemmerlie (1983) wrote a theory paper. The author defended the 1979 matching-law conclusions.
Critics had attacked the stats and methods. This paper says the attacks miss the mark.
What they found
The core finding: choice-experiment logic still holds. The old stats were fine.
Matching law conclusions do not need to be retracted.
How this fits with other research
Cox et al. (2017) extends the idea. They show Major League pitchers follow the same matching rule when they pick pitches.
Lanovaz et al. (2020) joins the method debate. They warn that too-tight tier rules can sink power in single-case studies.
Both papers keep the spotlight on sound design, just from different angles.
Why it matters
You can keep using matching-law analyses without apology. When you run choice probes or calculate response ratios, trust the math. If you consult multiple-baseline standards, borrow Lanovaz’s two-of-three rule to stay both safe and strong.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Plot your client’s response ratios across two concurrent tasks; check if the proportions match the reinforcement rates.
02At a glance
03Original abstract
Two papers have appeared recently, one by Mullins, Agunwamba, and Donohoe (1982) and one by Wearden and Burgess (1982), disagreeing with conclusions of my 1979 paper on choice experiments (Baum, 1979), the former on the basis of argument, the latter on the basis of additional data. Both these papers appear mistaken; the conclusions of my earlier paper still stand.
Journal of the experimental analysis of behavior, 1983 · doi:10.1901/jeab.1983.39-499