slopeDif: Conduct a slope difference test for three-way interaction...

In RobertGM111/mixedPowerSim: Functions for Testing and Simulating Three-Way Interactions in Mixed-Effects Models

Description

Calculates the expected value of the outcome variable (Y) at all combinations of high and low values (+/- 1SD) of X, Z, and W. Additionally, this function calculates the slope differences between all slope pairs.

Usage

slopeDif(x, )

Arguments

x	An object of class "lmerMod". The result of an lmer() function with a random intercept.
ZWithin	Logical. Required for all types. If TRUE, variable Z is simulated as a within-cluster variable. If FALSE, variable Z is simulated as a between-cluster variable.
WWithin	Logical. Required for all types. If TRUE, variable W is simulated as a within-cluster variable. If FALSE, variable W is simulated as a between-cluster variable.

Value

An object of class "slopetest" with the following components:

• Expected_Y: A data frame with the expected values of Y at high and low values of X, Z, and W.

• SlopeTests: A data frame representing slope tests as defined in Dawson and Richter (2006).

Examples

library(mixedPowerSim)
library(lme4)

nCluster <- 20
nObs <- 15
db0 <- .1
db1 <- .1
db2 <- .1
db3 <- .1
db4 <- .1
db5 <- .1
db6 <- .1
db7 <- .1
SDb0 <- 1
SDresid <- 5

simDat <- simMixedXYZ(nCluster = nCluster, nObs = nObs,
 db0 = db0, db1 = db1, db2 = db2,
 db3 = db3, db4 = db4, db5 = db5, db6 = db6, db7 = db7,
 SDb0 = SDb0, SDresid = SDresid, XWithin = TRUE, ZWithin = FALSE, WWithin = TRUE)

 mod <- lmer(Y~X * Z * W + (1|Cluster), data = simDat$simulatedData)

 slopeDif(mod)

RobertGM111/mixedPowerSim documentation built on Jan. 19, 2021, 12:34 a.m.