Based on faulty logic

Correlation ≠ Causation

The fact that a possible education outcome is associated with an intervention does not necessarily mean that the intervention caused the outcome. The association or correlation could instead be due to chance or some other underlying factor.

For example, learners who are taught using a new teaching strategy may simply be taught by more effective teachers than those learners who do not use that strategy. Therefore, those learners who receive the new intervention may appear to benefit, but the difference in outcomes could be because of the underlying quality of the teaching, rather than because of the intervention. We also know that new teaching methods tend to increase motivation and have a positive short-term effect. So, it is often difficult to say if better outcomes are actually due to a specific teaching method or just due to its newness.

BEWARE of claims that an intervention works because the intervention is correlated with the desired outcome.

REMEMBER: Do not assume that an outcome was caused by an intervention unless other reasons for an association have been ruled out by a fair comparison.