STR.7:6 - Bias-Annotation

Sponsors can prefer observations that preserve their idea. Specify the adverse and unresolved returns before the experiment and retain refusals, departures and missing observations at their actual meaning.

A low sponsor cost can conceal unpaid participant labour or exposure. Ask who bears the burden and who can decline. Small sample size limits the inference; a polished demonstration does not repair it.