top of page

Sample Size Calculator

Not sure which sample size calculation you need? Answer a few simple questions about your research question, and this calculator will guide you to an appropriate method.

​

This tool can be used for common medical and health research designs, including retrospective studies, observational studies, surveys, cross-sectional studies, cohort studies, case-control studies, and clinical trials.

​

Start by selecting the type of primary outcome you are studying. For example, your outcome may be numerical, such as blood pressure, age, BMI, cholesterol level, pain score, anxiety score, reaction time, hospital length of stay, or biomarker level. It may also be binary, such as disease vs no disease, event vs no event, responder vs non-responder, improved vs not improved, admitted vs not admitted, or complication vs no complication.

Step 1: Tell us about your outcome

Step 2: What are you comparing?

Use “two independent groups” when comparing two separate groups, such as exposed vs unexposed, disease vs no disease, treatment vs control, male vs female, responder vs non-responder, or two different cohorts.

Sample Size for Comparing Two Means

Use this when your primary outcome is numerical and you want to compare the average value between two independent groups.

Use standard deviations from previous studies, pilot data, published papers, or reasonable planning assumptions.

Sample Size for Comparing Two Proportions

Use this when your primary outcome is binary and you want to compare the percentage or rate between two independent groups.

Enter percentages from 0 to 100. For example, enter 30 for 30%.

Step 3: Choose your assumptions

Alpha is the cutoff used to decide statistical significance. Most studies use 0.05.

Power is the probability of detecting the expected difference if it truly exists. Many studies use 80% or 90%.

Use 1 for equal group sizes. For example, 1 means 1:1 allocation.

​

​

bottom of page