Last updated: May 2023.
Authors:
- Zarni Htet
- Kristen Hunter
- Luke Miratrix
- Kristin Porter
For randomized controlled trials (RCTs) with a single intervention being measured on multiple outcomes, researchers often apply a multiple testing procedure (such as Bonferroni or Benjamini-Hochberg) to adjust
Please see the vignettes for examples of how to use this package.
More details about the package:
- For a full package description, including a detailed technical appendix, see: https://arxiv.org/abs/2112.15273.
- See also the shiny app: https://public.mdrc.org/pump/.
- For a presentation about PUMP, see: https://github.com/kristenbhunter/presentations/tree/master/NCI2022.
We provide below one example of using PUMP to calculate a minimium detectable effect size (MDES). The user specifies the RCT design and model (d_m), the multiple testing procedure (MTP, in this case Holm), the target power (0.8), and the type of power desired (individual power for outcome 1). The user also specifies a variety of design and model parameters, such as the number of outcomes, sample sizes at different levels, variation explained by covariates, etc.
m <- pump_mdes(
d_m = "d3.2_m3fc2rc", # choice of design and analysis strategy
MTP = "HO", # multiple testing procedure
target.power = 0.80, # desired power
power.definition = "D1indiv", # power type
M = 5, # number of outcomes
J = 3, # number of schools per block
K = 21, # number districts
nbar = 258, # average number of students per school
Tbar = 0.50, # prop treated
alpha = 0.05, # significance level
numCovar.1 = 5, # number of covariates at level 1
numCovar.2 = 3, # number of covariates at level 2
R2.1 = 0.1, R2.2 = 0.7, # explanatory power of covariates for each level
ICC.2 = 0.05, ICC.3 = 0.4, # intraclass correlation coefficients
rho = 0.4 ) # how correlated outcomes are