Your portal for simplifying the power analysis process and determining required sample sizes for researchers, physicians, students, and academics. Select the appropriate statistical test, plan your sample size, and access core concepts all on a single page.
Power analysis evaluates the probability of detecting a true effect. Insufficient sample size may miss real differences.
Informative cards and short explanations
The probability of detecting a true effect. In many studies, 80% or 90% is targeted.
The probability of Type I error, usually 0.05. Lower alpha requires larger samples.
Represents the magnitude of a difference or association. Highly valuable for clinical interpretation.
Too small causes weak conclusions; too large wastes time and cost.
All our power analysis tools are just a click away
Compare the means of two independent groups.
Compare before-after (dependent) measurements in the same subjects.
Compare two independent groups non-parametrically (medians).
Compare two dependent measurements non-parametrically.
Compare the means of three or more independent groups.
Compare 3 or more repeated measurements of the same subjects.
Test association and goodness-of-fit for categorical variables.
Perform 2x2 proportion analysis on very small samples.
Compare success or event rates in two independent groups.
Measure the strength of the linear relationship between two continuous variables.
Predict a continuous outcome (R²) with multiple predictors.
Predict a binary outcome using independent variables.
A guided wizard to quickly find the most suitable analysis
Answer the questions below to find the most appropriate statistical test for your research.
Explanatory content area within the homepage
Ideally before starting the study, as an a priori analysis.
It is debated; effect size and confidence intervals are often more informative.
You can use the literature, pilot data, or a clinically meaningful minimum difference.
Yes. Many core modules can be developed in PHP and JavaScript.
