Reading the evidence
A false negative occurs when a test fails to detect an effect or condition that is present. Small samples, noisy measurement or strict thresholds can make detection unlikely.
Make it click
An alarm can miss a small or poorly placed fire. Silence does not prove safety when the device or test lacks enough sensitivity.
Now put that into markets
A market study can produce an inconclusive result even when a modest effect exists. Statistical power describes detection probability under specified alternatives.
Type II error depends on effect size, variability, sample and threshold. Failing to reject a null is not the same as proving no effect.
Show intervals and detectable effect sizes. Distinguish evidence of absence from absence of adequate evidence.
Making every test more sensitive raises false alarms, while searching only for larger effects can miss economically useful ones.
Takeaway
