Tamanho amostra poder
- Created by
- Renato Passos, Eng. de Software
- Reviewed by
- Renato Passos, Eng. de Software
Last updated: Apr 18, 2026
Formula
n = 2((zα+zβ)/δ)²
About this calculator
This calculator determines the sample size required for a statistical test based on the test's power. The power of a test is the probability of detecting an effect when it exists. The formula used here, n = 2·((z_α + z_β)/d)², considers type I (α) and type II (β) errors, represented by the critical values z_α and z_β, and the effect size (d).
The formula works by considering z_α as the critical value from the standard normal distribution corresponding to the type I error (usually 0.05), z_β as the critical value related to the test power (1 - β), and d as the effect size. The result, n, is the number of samples needed in each group to detect the effect with the specified power.
It's crucial to use this calculator in contexts where detecting effects is vital, such as in clinical or market studies, where precision in sampling can mean the difference between accurate and inaccurate conclusions. However, care must be taken with estimates of effect size and chosen error levels, as they directly impact the required sample size.
When interpreting results, remember that a larger sample size generally provides more precise estimates but also increases the cost and time needed for data collection. Therefore, it's essential to balance these factors according to the study's objectives and available resources.
Frequently asked questions
What is test power?
Test power is the probability of detecting an effect when it actually exists. It's a measure of the test's ability to identify a significant effect.
How to choose α and β values?
The values of α (type I error) and β (type II error) should be chosen based on study objectives and error consequences. Conventionally, α is set at 0.05 and β at 0.20.
What is effect size (d)?
Effect size (d) measures the magnitude of the effect to be detected. It's crucial for calculating the required sample size.
Can I use this calculator for any type of study?
This calculator is suitable for studies involving statistical tests comparing groups. However, parameter choice must be done carefully, considering each study's specifics.
What if I don't know the effect size?
If you don't know the effect size, it's recommended to conduct a pilot study to estimate it or consult existing literature for a realistic estimate.