Randomization Schemas

This project is a full demonstration of designing an experiment, with an application in Shiny.
code
statistics
designs
college
Author

Matthias Quinn

Published

September 18, 2020

GOAL

Produce the same schema that you have, which represents the desired study design. The input to your programs should be the parameters of # of sites, # of subjects/site, a randomization ratio, and various stratification levels.

RESULT

A reactive application that creates a design schema, given the researchers design parameters.

This project was finished in about 5 weeks.

Let’s say we’re a researcher interested in deciding which treatment to give to our target audience. Since we want to make sure we follow proper guidelines, we decide to reach out to a friend of ours, a statistician, for advice.

She says that a good practice would be to create a credible, experimental design to find what we’re looking for. This project is essentially a reproducible example of our goals with the following condition:

  1. Write a completely randomized design schema for S subjects at T sites in blocks of B where randomization is N:D in terms of treatment to control.

Running Example:

30 subjects at site one in blocks of 6 where the randomization of treatment to control is 1:1

Solution:

Setting a seed is necessary for reproducibility.

Also, having explainable code was a requirement for this project, as the client may have wanted to know how to make changes in the future.