A valid evaluation of the theory of multiple intelligences is not yet possible: Problems of methodological quality for intervention studies

39 articles, 3,009 students, 14 countries

Pre-school to high-school students of diverse achievement levels

Systematic review plus attempted meta-analysis of MI-inspired classroom interventions. Inclusion required a quantitative estimate of impact on academic performance (reading, maths or science) using a pre-post design with a control group - i.e. the authors set a LOW bar and the literature still could not clear it. The headline result is a statement about the evidence base, not about MI's efficacy.

The studies had important methodological flaws - small samples, absence of active control groups - and reported insufficient information about the outcome instruments and about what was actually done during training. Signs of publication or reporting bias were present. The authors conclude that a valid evaluation of MI-based instruction is NOT YET POSSIBLE. This is the difference between 'tested and failed' and 'never operationalised well enough to test', and MI is the latter.

Medium. The pooled studies are largely uncontrolled or passively controlled classroom comparisons in which selection into MI classrooms plausibly tracks family and child characteristics; nothing in the literature is genetically informative.