Interactive tool for calculating required sample sizes across different study designs and sampling methods. Visualize sampling strategies and understand design effects.
Select your test type
Probability of detecting an effect if it exists (typically 80%)
or use a preset
Choosing a preset will auto-fill N. You can still edit it.
Configure finite population correction and design-specific options.
Choose design and extraction
Single homogeneous population
Sample within subgroups
Sample groups, then all within
Strata, then clusters within
Proportional allocates by stratum size; Equal assigns n/S per stratum.
Stratification can reduce required N when strata explain outcome variation. % reduction approximates this gain.
Design effect: DEFF = 1 + ((m − 1) × ρ), where m is average cluster size and ρ is ICC.
Proportional allocates by stratum size; Equal assigns n/S per stratum.
Combined DEFF ≈ cluster DEFF × (1 − stratification %).
Equal probability for all units
Select every kth unit
Based on your study parameters
Current Design: With N = 73, you have 57.7% power to detect an effect size of 0.50 at α = 0.05.
APA-formatted summary of your sampling design
| Parameter | Value | Description |
|---|---|---|
| Test Type | — | — |
| Effect Size (d) | — | — |
| Significance Level (α) | — | — |
| Statistical Power (1-β) | — | — |
| Test Direction | — | — |
| Sampling Design | — | — |
| Extraction Method | — | — |
| Required Sample Size | — | — |
Sample size calculated using power analysis for mean differences. Design effects and finite population corrections applied where specified.
Reproduce this analysis in R using the pwr package
Copy this code to reproduce the analysis in R (pwr). Python support is coming soon.
pwr; minor differences may occur for the reasons above.
install.packages()library()