In choosing a design, the investigator should carefully consider the research question, the context of the research, and the nature of the intervention. This figure presents three key questions to guide the selection process among four major design categories: randomized controlled trial (RCT), individually randomized group-treatment (IRGT) trial, parallel group- or cluster-randomized trial (GRT), and stepped-wedge group- or cluster-randomized trial (SW-GRT).
a If any of the intervention is delivered through physical or virtual groups, or through shared intervention agents who each work with multiple participants, positive intraclass correlation can develop over the course of the trial.
b There may be logistical reasons to randomize groups or clusters or it may not be possible to deliver the intervention to individuals without substantial risk of contamination.
c There may be political or logistical reasons to roll out the intervention to all groups or clusters before the trial is over.
Adapted from Murray DM, Taljaard M, Turner EL, George SM. Essential Ingredients and Innovations in the Design and Analysis of Group-Randomized Trials. Annual Review of Public Health. 2020;41:1-19. PMID31869281.
Learn more about choosing the design
The familiar methods for sample size estimation used in randomized clinical trials require modifications for use in parallel group- or cluster-randomized trials (GRTs), individually randomized group-treatment (IRGT) trials, stepped wedge group- or cluster-randomized trials (SWGRTs), and group or cluster regression discontinuity designs (GRDDs) to account for the positive intraclass correlation (ICC) expected among members of the same group or cluster (Campbell and Walters, 2014; Donner and Klar, 2000; Eldridge and Kerry, 2012; Hayes and Moulton, 2017; Moerbeek and Teerenstra, 2016; Murray, 1998; ). The specific modifications will depend on the design of the study, the expected distribution of the outcome variable, and the analytic plan.
The terminology and methods used to estimate sample size for GRTs and IRGT trials are based on work published in the first textbook on the design and analysis of parallel GRTs (Murray, 1998) and updated in subsequent articles (Janega et al., 2004; Murray et al., 2007; Pals et al., 2008). The terminology and methods used for time to event outcomes in GRTs are based on the Additive Hazards Mixed Model (Blaha et al., 2022). The terminology and methods used to estimate sample size for SWGRTs are based on several important articles (Hemming et al., 2020; Hooper et al., 2016; Hughes et al., 2024; Kasza and Forbes, 2019b; Kasza et al., 2019a; Kasza et al., 2020; Kenny et al., 2022). The terminology and methods used to estimate sample size for GRDDs are based on two key articles (Schochet 2009; Pennell et al., 2011).
As noted above, it is important for investigators to become familiar with the material on this website. It is even more important that they collaborate with a methodologist familiar with these issues, particularly in the development of the study design, analytic plan, and sample size calculations. Investigators may find it difficult to work through the sample size calculator on their own and may have a better experience by turning that task over to their methodologist.
Tips for Using the Sample Size Calculators
- Information will be saved for each step only if you click the "Continue" button. Any information you save will be retained unless it is overwritten by a subsequent selection or entry.
- Use the "Back" button to return to the previous step.
- Click on any of the completed steps in the list on the left side of each page to go back to that step. Change your entry if you wish, and after clicking "Continue," you can move to any other step, including "Analysis."
- You can repeat this exercise as often as you wish to compare different designs and/or analytic plans or change your parameter estimates.
- You can print or download the results from each exercise.
There are nine steps for each sample size calculation. You will be asked to specify the type I error rate and desired power for the test of your intervention effect, the expected distribution of your outcome variable, and the design and analytic plan for your trial. You will also be asked to provide estimates of the parameters needed to estimate the sample size required for your study. Please note that this sample size calculator assumes the same variance components, ICCs, and group sizes in both study conditions; those assumptions are appropriate for most parallel group-randomized trials (GRTs).
Launch the Parallel GRT Sample Size Calculator
Additional Resources
For more information, please review a list of recent publications on sample size estimation methods for parallel GRTs or watch a Methods: Mind the Gap webinar, which demonstrates how to use the parallel GRT calculator.
Use this calculator to estimate sample size for an individually randomized group-treatment (IRGT) trial with clustering in only one arm. If you expect to have clustering in both arms, you can use the GRT Sample Size Calculator, because GRTs always have clustering in both arms. Please note, however, that the GRT Sample Size Calculator assumes the same variance components, ICCs, and group sizes across study conditions, which may not always be appropriate for an IRGT trial with clustering in both arms (
).
There are eight steps for each sample size calculation for an IRGT trial. You will be asked to specify the type I error rate and desired power for the test of your intervention effect, the expected distribution of your outcome variable, and the design and analytic plan for your trial. You will also be asked to provide estimates of the parameters needed to calculate the sample size required for your study.
Launch the IRGT Sample Size Calculator
Additional Resources
For more information, please review a list of recent publications on sample size estimation methods for IRGT trials or watch a Methods: Mind the Gap webinar, which provides useful information on their design, analysis, and sample size estimations.
There are seven steps for each sample size calculation. You will be asked to specify the type I error rate and desired power for the test of your intervention effect, the expected distribution of your outcome variable, and the design and analytic plan for your trial. You will also be asked to provide estimates of the parameters needed to estimate the sample size required for your study. Please note that this sample size calculator assumes the same variance components, ICCs, autocorrelations, and group sizes in all sequences.
Launch the SWGRT Sample Size Calculator
Additional Resources
For more information, please review a list of recent publications on sample size estimation methods for SWGRTs or watch a Methods: Mind the Gap webinar, which provides an overview of statistical models used in the design and analysis of SWGRTs.
There are seven steps for each sample size calculation. You will be asked to specify the type I error rate and desired power for the test of your intervention effect, the expected distribution of your outcome variable, and the design and analytic plan for your trial. You will also be asked to provide estimates of the parameters needed to estimate the sample size required for your study. Please note that this sample size calculator assumes the same variance components, ICCs, number of groups, and group sizes in both study conditions. In addition, this calculator utilizes all observations to either size of the cutoff and does not allow for specification of a smaller bandwidth about the cutoff.
Launch the GRDD Sample Size Calculator
Additional Resources
For more information, please review a list of recent publications on sample size estimations methods for RDDs and GRDDs or watch a Methods: Mind the Gap webinar, which provides an overview of statistical models used in the design and analysis of RDDs.
Suggested citation: Research Methods Resources: National Institutes of Health. [Accessed Month Day, Year]. Available from: https://researchmethodsresources.nih.gov/.