power2L: power2L

In bangtle/PowerRTCRT: Calculates the power for Randomized Trials (RT) and Cluster Randomized Trials (CRT) models.

Description

This function returns the power for two-level cluster randomized trials designs with treatment at level 2 (cluster level). It addresses the following models:

• Main treatment effects with no additional covariates.

• Main treatment effects with covariate(s) at level 1 only.

• Main treatment effects with covariate(s) at level 2 only.

• Main treatment effects with covariate(s) at levels 1 and 2.

• Cluster-level moderator effects with no additional covariates.

• Cluster-level moderator effects with covariate(s) at level 2 only.

• Individual-level moderator effects with no additional covariates.

• Individual-level moderator effects with covariate(s) at level 1 only.

Usage

power2L(n, J, delta, rhoL2, numcovL1 = 0, numcovL2 = 0, R2L1 = 0,
  R2L2 = 0, modL1 = FALSE, modL2 = FALSE, alpha = 0.05)

Arguments

n	Number of units in level 1 (individual level).
J	Number of units in level 2 (cluster level).
delta	Standardized effect size.
rhoL2	Intraclass correlation (ICC) at level 2.
numcovL1	Number of covariate(s) at level 1. Default value is 0.
numcovL2	Number of covariate(s) at level 2. Default value is 0.
R2L1	R-squared at level 1. Default value is 0.
R2L2	R-squared at level 2. Default value is 0.
modL1	Existence of moderator at level 1. Default value is FALSE.
modL2	Existence of moderator at level 2. Default value is FALSE.
alpha	Probability of type I error. Default value is 0.05.

See Also

PowerUpR

Examples

# All arguments are specified
power2L(n=100,J=40,delta=0.2,rhoL2=0.3,numcovL1=1,numcovL2=0,R2L1=0.2,R2L2=0,modL1=TRUE,modL2=FALSE,alpha=0.05)

# Unspecified arguments take their default values
power2L(n=100,J=40,delta=0.2,rhoL2=0.3)

bangtle/PowerRTCRT documentation built on May 11, 2019, 6:19 p.m.