About this applet
This web applet calculates the minimum sample size necessary for a SMART with an end of study binary or continuous outcome, in which the primary aim is a comparison of any two adaptive interventions starting with different interventions.
How can a behavioral scientist use this applet?
Behavioral intervention scientists can use this code to calculate the sample size for SMART with a continuous or binary end-of-study outcome.
What method does this applet implement?
This code implements sample size formulae based on a test of the null hypothesis that the difference in means (continuous outcome) or probabilities of success (binary outcome) between two adaptive interventions is zero.
Related References
1. Kidwell, K. M., Seewald, N. J., Tran, Q., Kasari, C., & Almirall, D. (2018). Design and analysis considerations for comparing dynamic treatment regimens with binary outcomes from sequential multiple assignment randomized trials. Journal of applied statistics, 45(9), 1628-1651.