Is Developing Scientific Thinking All About Learning to Control Variables?

Small sample of academically low-performing urban sixth graders (exact n not obtained)

US urban middle school; academically low-performing grade 6 students (~age 11-12).

A deliberate counterweight to the Klahr/Chen control-of-variables programme. Students engaged in a multi-session inquiry task; the manipulation was a minimal verbal suggestion to focus the investigation on the role of a single factor. Outcome is the students' own experimentation behaviour in the task — a trained-task measure, researcher-designed, scored by the researchers. Small, short, and not an efficacy trial of anything schools could adopt; its value is conceptual, not as an effect estimate. The argument matters for `outcome_class` discipline: if a one-sentence hint reproduces much of what an explicit CVS training programme achieves, then "CVS acquisition" is a performance effect about what students choose to attend to, not the acquisition of a general reasoning capacity — which is precisely why CVS gains should never be scored as far transfer.

Simply suggesting to low-performing sixth graders that they investigate one factor at a time produced a significant change in their control of the other variables. Kuhn & Dean argue that variable control has come to stand in for scientific thinking as a whole, that it is only one component, and that what develops is the disposition and metacognitive control to deploy a strategy students often already possess rather than the strategy itself. Read against Chen & Klahr (1999), the pair frames the boundary of this literature: the trainable thing is narrow and situation-cued, and no measured outcome leaves the task.

Low relevance — within-task behavioural manipulation, but the design is too small and uncontrolled to support causal generalisation anyway.

Not directly replicated as reported here. The broader claim — that inquiry-task performance is highly cue-dependent — is consistent with the classroom reversal literature already in this archive (cvs-retrieval-2024-classroom-reversal) and with Klahr & Nigam (2004).