Propensity Score Calculator
Subjects receiving treatment/intervention
Subjects in control group
Assumed equal variance
Results
Intercept (β₀): 0.00
Covariate Coefficient (β₁): 0.00
Propensity Score: 0.00
Propensity Score Examples
| Covariate Value | Propensity Score | Interpretation |
|---|
Understanding Propensity Score Calculation
What is Propensity Score Matching?
Propensity score matching is a statistical technique used in observational studies to estimate the effect of a treatment by accounting for covariates that predict receiving the treatment. This method helps reduce selection bias by creating comparable groups between treated and control subjects.
Propensity Score Formula
The propensity score is calculated using logistic regression:
e(X) = P(T=1|X) = 1 / (1 + exp(-(β₀ + β₁X)))
| Variable | Description | Unit |
|---|---|---|
| β₀ | Intercept term | Unitless |
| β₁ | Covariate coefficient | Unitless |
| X | Observed covariate | Variable-specific |
Practical Examples
Example 1: Medical Treatment Study
- 200 treated patients (mean age 45), 180 controls (mean age 50)
- Calculated β₁ = 0.82, propensity score of 0.72 for 48-year-old patient
How to Use This Calculator
- Enter sample sizes for both groups
- Input covariate means for each group
- Specify standard deviation
- Enter X value for prediction
Key Influencing Factors
- Covariate distribution overlap
- Sample size ratio
- Measurement precision
- Model specification
- Standardized mean differences
- Variance homogeneity
Can I use multiple covariates?
This calculator demonstrates single-covariate analysis. Real-world applications typically use multiple covariates.
Related Resources