Four principles of taking statistical uncertainty into account when estimating and reporting effects are discussed. Means of implementing the principles, ways in which the principles are violated in practice, implications for the use of multiple methods, effect size, estimation techniques, and random and nonrandom certainty are described. (TJH)