StatrivaPlan the study before collecting data with a defensible sample size.
Sample size calculations should follow the planned design, outcome, effect of interest, significance level, power target, and practical constraints. I help turn those choices into a transparent statistical justification.
The number depends on the design
A sample-size calculation is not a single formula. The correct approach depends on what you want to estimate or detect.
Continuous, binary, count, time-to-event, proportion, or another outcome structure.
Use a scientifically defensible effect size rather than choosing a number simply because it produces a convenient sample.
Groups, repeated measurements, attrition, clustering, predictors, and other features can change the required sample.
Receive a clear calculation and explanation suitable for a protocol, proposal, methods section, or ethics submission.
Examples of study types
A calculation you can explain
Assumptions
Inputs such as alpha, target power, expected effect, variability, allocation, and design assumptions are made explicit.
Calculation
The appropriate calculation is performed for the proposed analysis, rather than applying a generic sample-size rule.
Written rationale
The result is explained in clear language so you can understand and defend the basis for the planned sample.
Three steps before recruitment
Describe the design
Share the study objective, primary outcome, groups, planned analysis, and any practical recruitment constraints.
Choose defensible inputs
Discuss the effect size and other assumptions, including how they are supported by prior evidence or pilot information.
Document the calculation
Receive the numerical result plus a transparent explanation of the assumptions and interpretation.
Still deciding what analysis you will use?
That matters for sample size. If the statistical model is not settled yet, start with the research question and study design rather than calculating a number too early.
Discuss your study