To match or not to match? Methodological issues in autism-related research.
Match on the skill you plan to study, then use regression, not just IQ.
01Research in Context
What this study did
Christopher and colleagues wrote a how-to paper. They looked at common ways researchers pair autistic kids with non-autistic controls.
The team warned that matching only by IQ can hide real differences. They urged using task scores and regression instead.
What they found
The paper says: stop relying on IQ-alone matches. Add tests that check the skill you plan to study.
Use regression to keep small leftover differences from skewing your results.
How this fits with other research
Ganz et al. (2004) said the same thing in the same year. Both papers tell us to set stricter rules before calling groups 'matched.'
Flapper et al. (2013) later built on the idea. They showed how to report effect sizes and variance ratios to prove groups are truly equal.
Hastings et al. (2001) took the old IQ-only path. Christopher’s paper quietly explains why studies like that can miss key cognitive gaps.
Why it matters
Next time you run a group study, test the exact skill first. Plug those scores, plus IQ, into a regression model. Your stats will control for leftover differences instead of pretending they don’t exist. Cleaner data, clearer answers, better interventions.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a short pre-test of your target skill and include it as a covariate in your next group analysis.
02At a glance
03Original abstract
Studies of autism typically adopt a factorial matched-groups design aimed at eliminating nonspecific factors such as mental retardation as explanations of performance on experimental tasks. This paper reviews the issues involved in designing such studies and interpreting their results and suggests that the best approach to matching may be to equate performance on carefully designed control tasks. However, we also argue that the interpretation of such studies is often complicated by the fact that associations between background measures and experimental task performance are not clear. Consequently, we also advocate the use of regression techniques that allow the researcher to determine the factors that relate to task performance and to assess the extent of group differences on the task of interest having taken these factors into account.
Journal of autism and developmental disorders, 2004 · doi:10.1023/b:jadd.0000018078.82542.ab