Logistic Regression Power Calculator
This advanced calculator determines the statistical power for logistic regression models. Statistical power is the probability that your test will detect an effect when there is one. Use this tool to plan your study, determine required sample sizes, or calculate achieved power.
Calculator Inputs
Enter your study parameters below. All fields are required for accurate power calculation.
Enter the year your model applies to
Total number of observations in your study
Probability of the event occurring (between 0.01 and 0.99)
Total predictors in your logistic regression model
Expected odds ratio for your primary predictor (must be >1)
Significance level
Variance in the primary predictor explained by other predictors (0–0.95)
Results
Your logistic regression power analysis results appear below.
Statistical Power
—
Probability of detecting the effect (80%+ is recommended).
Required Sample Size (for 80% power)
—
Minimum N needed to achieve 80% power with your parameters.
Adjusted Effect Size (f²)
—
Cohen’s f² after adjustment for other predictors.
Power Curve
Statistical Power Guidelines
Statistical power is crucial for valid research conclusions. Below are international standards for power interpretation:
- Adequate Power: 0.80 or higher (80% chance to detect true effect)
- High Power: 0.90 or higher (often used in clinical trials)
- Low Power: Below 0.80 (increased risk of Type II error)
- Insufficient: Below 0.50 (high risk of missing true effects)
| Power Level | Interpretation | Recommended Use |
|---|---|---|
| 0.90 – 1.00 | Excellent power | Clinical trials, regulatory studies |
| 0.80 – 0.89 | Adequate power | Most research studies, publication standard |
| 0.70 – 0.79 | Moderate power | Exploratory research, pilot studies |
| Below 0.70 | Low power | Preliminary analysis only |
Use the Least Squares Linear Regression Calculator for precise linear analysis, or explore the full Regression Calculator category to access all regression tools in one place.