Providing Instruction Based on Students' Learning Style Preferences Does Not Improve Learning

125 fifth-graders (118 with complete data); interaction tested on the categorically classified subset

Fifth-graders aged 10-11, rural Pennsylvania, United States

The first Pashler-compliant test of the meshing hypothesis on SCHOOL-AGED children, and therefore the single most important study for an ages 4-18 archive. Learning style measured with the Rundle and Dunn 'Learning Style: The Clue to You!' inventory for ages 10-13. Mixed design: every student received both listening and reading instruction using Gates-MacGinitie passages, with mode and test form counterbalanced, and all comprehension questions were identical and text-based.

The critical learning-style x instructional-mode interaction was not significant, F(1,32) = 1.16, p = .29. What did emerge was a main effect: children classified as visual learners outperformed those classified as auditory learners, F(1,32) = 12.92, p = .001, eta-squared = 0.29 - and they did so on BOTH listening and reading comprehension, which is the anti-crossover pattern. Authors conclude matching had no effect on achievement and that resources spent diagnosing styles are wasted.

Low for the causal claim. The design is within-subject on mode with counterbalancing, so the interaction is identified without needing groups to be exchangeable. The main effect (visual-preference children better at everything) is confounded and should be read as preference tracking verbal ability, not as a modality strength.