Simulate cluster-randomized trial data with binary outcomes

Simulate cluster-randomized trial data with binary outcomes

Usage

genBinCRT(
  m = c(1, 1),
  m_alloc = c(0.5, 0.5),
  n = c(1, 1),
  n_cv = c(0, 0),
  p_vec = c(0.5, 0.5),
  rho = c(0, 0)
)

Arguments

m

Number of clusters.

m_alloc

Allocation ratio of clusters per arm. Default is c(0.5, 0.5).

n

Mean size of each cluster.

p_vec

Probabilities of event for control and treatment groups.

rho

Intraclass correlation coefficient. Default value is 0.

Value

Simulated binary outcomes represented as an n x 4 matrix with the columns encoding the treatment arm, cluster, individual, and response.

Author

Lee Ding lee_ding@g.harvard.edu