The core principle asserts that statistical power and Type I error rates in hypothesis testing are non-linear functions of sample size ($N$), fundamentally altering the sensitivity of p-values to effect magnitude. This concept operates within inferential statistics as a formal mechanism describing how decreasing $n$ increases variance in test statistics, thereby diminishing the likelihood of rejecting a false null hypothesis while maintaining strict control over alpha levels under asymptotic or finite-sample theoretical frameworks.
Prerequisite chain context: requires Null Hypothesis Formulation for Normality.