Why is 80% the standard power threshold?

Jacob Cohen (1988) argued that a Type II error (failing to detect a real effect) is 4x less serious than a Type I error (false positive). With α = 0.05, setting power = 0.80 (β = 0.20) maintains a 4:1 β:α ratio. Applied journals increasingly require 80–90% power justification in methods sections.

What is G*Power and how is it used?

G*Power is free software (Faul et al., 2007) widely used for power analysis across 50+ statistical tests including t-tests, ANOVA, chi-square, correlations, and regression. It solves for any one of: sample size, effect size, alpha, or power given the other three. It is the standard reference tool for sample size planning in published research.

Statistical Power Analysis for Research Studies

How It Works

Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations

Note: Interpret the statistical power result against the clinical thresholds and context described above.

How to Use

Enter the values for the patient or scenario you are assessing. Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations. Use the statistical power result to inform your clinical assessment.

Glossary and Definitions

Statistical power: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.
Power analysis: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.
Sample size planning: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.
Type II error: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.
Effect size: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.
Hypothesis testing: A key concept in Statistical Power: Calculate statistical power (1–β) given sample size, effect size (Cohen's d), and significance level α. Or solve for required sample size given target power. Supports one-sample, two-sample, and paired t-test configurations.