如何计算Statistical Power
learn.whatIsHeading
Statistical power (1−β) is the probability of correctly detecting a true effect. Power of 0.80 (80%) is the conventional minimum — a 20% chance of missing a real effect (Type II error).
分步指南
- 1Power increases with: larger n, larger effect size, higher α
- 2Plan sample size to achieve 80% power before collecting data
- 3Power = Φ(|d|√(n/2) − z_α)
例题解析
输入
d=0.5 · n=50/group · α=0.05
结果
Power ≈ 70% — underpowered
Need ~64/group for 80% power
准备好计算了吗?尝试免费的 Statistical Power 计算器
自己尝试一下 →