Common research practice backfires in statistical analysis, study warns

Researchers using a standard statistical technique to control for baseline measurements are inadvertently distorting their results, new simulations show. The finding matters because organizations rely on these analyses to evaluate program effectiveness, policy impact, and treatment outcomes—meaning flawed methodology could be driving millions in misdirected funding decisions.

Originaltitel: Dangers of including outcome at baseline as a covariate in latent change score models: Results from simulations and empirical re-analyses