G*Power Sample Size Calculator

An essential tool for researchers to perform an a priori power analysis and determine the optimal sample size for their studies.

Power Analysis Visualization

Chart showing the relationship between Effect Size, Statistical Power, and the required Sample Size.

What is a Sample Size Calculation Using G*Power?

A sample size calculation, often performed using software like G*Power, is a critical step in research design called an a priori power analysis. Its purpose is to determine the minimum number of participants or observations needed to have a reasonable chance of detecting a true effect, if one exists. This process balances the need for statistical significance against the practical constraints of collecting data. Inadequate sample sizes can lead to studies that are “underpowered,” meaning they have a low probability of finding a real effect, while excessively large samples waste resources. This calculator helps you **calculate sample size using G*Power** principles for one of the most common statistical tests: the independent samples t-test.

The Formula to Calculate Sample Size

While G*Power uses complex algorithms, the core logic for an independent two-sample t-test is based on the non-central t-distribution. A simplified formula that provides a good approximation for the sample size per group (n) is:

n = 2 * ( (Z_α/2 + Z_β) / d )²

This formula highlights how the required sample size is influenced by the desired alpha, power, and the expected effect size. This calculator uses a more precise iterative method to find the exact sample size.

Variables in the Sample Size Calculation

Variable	Meaning	Unit	Typical Range
n	Sample size per group	Count (Participants)	Varies (10 – 1000+)
Zα/2	The critical value from the standard normal distribution for the given alpha level (for a two-tailed test).	Standard Deviations	1.96 (for α=0.05), 2.58 (for α=0.01)
Zβ	The critical value from the standard normal distribution for the desired power (1-β).	Standard Deviations	0.84 (for Power=0.80), 1.28 (for Power=0.90)
d	Cohen’s d (Effect Size)	Standard Deviations	0.2 (small), 0.5 (medium), 0.8 (large)

Practical Examples

Example 1: Medium Effect Size

A psychologist wants to test a new therapy to reduce anxiety. They expect a medium effect size based on prior research.

• Inputs: Effect Size (d) = 0.5, Alpha (α) = 0.05, Power = 0.80
• Units: Not applicable (standardized)
• Results: The calculator recommends a total sample size of 128 (64 participants in the therapy group and 64 in the control group).

Example 2: Small Effect Size

A market researcher is testing a subtle change in packaging to see if it increases sales. They anticipate a small effect.

• Inputs: Effect Size (d) = 0.2, Alpha (α) = 0.05, Power = 0.80
• Units: Not applicable (standardized)
• Results: To reliably detect such a small effect, the calculator recommends a much larger total sample size of 788 (394 per group).

How to Use This G*Power Sample Size Calculator

Follow these steps to determine the appropriate sample size for your research:

1. Select Test Family: Currently, this calculator is designed for t-tests comparing two independent means.
2. Enter Effect Size (Cohen’s d): Estimate the magnitude of the effect you expect to find. Use values from previous research, a pilot study, or use benchmarks (0.2 for small, 0.5 for medium, 0.8 for large).
3. Set Alpha (α) Probability: This is your threshold for statistical significance, typically 0.05.
4. Set Power (1-β): This is your desired probability of detecting a true effect. 0.80 (or 80%) is a widely accepted standard.
5. Click “Calculate”: The tool will instantly provide the total required sample size, the sample size per group, and other key metrics.
6. Interpret Results: The primary result is the total number of participants you should aim to recruit for your study to have the desired power.

Key Factors That Affect Statistical Power

Several factors interact to determine the statistical power of a study, and consequently, the required sample size.

Effect Size:

This is the magnitude of the difference or relationship you’re trying to detect. Larger effects are easier to find and require smaller sample sizes. A small effect requires a large sample size to be detected confidently.

Sample Size (N):

The number of participants in your study. A larger sample size generally increases statistical power. This calculator is designed to find the optimal sample size.

Significance Level (Alpha, α):

The probability of making a Type I error (rejecting a true null hypothesis). A stricter (lower) alpha level (e.g., 0.01 instead of 0.05) decreases power, requiring a larger sample size to compensate.

Variability of the Data:

Higher variance (or standard deviation) within your samples makes it harder to detect a true effect, thus decreasing power. You can sometimes reduce variability by using more precise measurement tools or a more homogeneous sample.

One-Tailed vs. Two-Tailed Test:

A one-tailed test (which predicts the direction of the effect) is more powerful than a two-tailed test. However, two-tailed tests are more common and conservative as they can detect an effect in either direction.

Study Design:

Within-subjects designs (like paired t-tests) are generally more powerful than between-subjects designs because they control for individual participant variability.

Frequently Asked Questions (FAQ)

1. What is statistical power?

Statistical power is the probability that a test will correctly detect a true effect when one exists. It’s like the sensitivity of your study. A power of 0.80 means you have an 80% chance of finding a statistically significant result if the effect you’re looking for is real.

2. What if I don’t know my effect size?

This is a common challenge. You can: 1) Look at similar studies in your field to get an estimate. 2) Conduct a small pilot study to calculate a preliminary effect size. 3) Default to a conventional “medium” effect size (d=0.5), but acknowledge this is an assumption.

3. Why is 0.80 a common choice for power?

It represents a consensus-based balance. It implies a 20% risk of a Type II error (β = 0.2), which is four times the typical risk of a Type I error (α = 0.05). This reflects the view that failing to detect a real effect is serious, but less so than falsely claiming an effect exists.

4. What is the difference between alpha and beta?

Alpha (α) is the probability of a Type I error (a false positive: finding an effect that isn’t there). Beta (β) is the probability of a Type II error (a false negative: failing to find an effect that is there). Power is calculated as 1 – β.

5. Does increasing my sample size always increase power?

Yes, up to a point. Increasing the sample size is one of the most direct ways to increase statistical power. However, there are diminishing returns. The power chart shows that the gain in power becomes smaller with each additional participant after a certain point.

6. What is Cohen’s d?

Cohen’s d is a standardized measure of effect size. It represents the difference between two group means, expressed in terms of their pooled standard deviation. This standardization allows for the comparison of effect sizes across different studies and measures.

7. Can I use this calculator for ANOVA or regression?

No, this specific calculator is designed for an a priori power analysis for a two-group independent samples t-test. Power calculations for ANOVA (F-tests) or regression analyses involve different formulas and more parameters (e.g., number of groups, number of predictors).

8. What does “a priori” power analysis mean?

“A priori” means “from the former” in Latin. An a priori power analysis is conducted *before* you collect data, during the design phase of a study. This is the most common and recommended type of power analysis, as it helps you plan your resources effectively.