我已经使用lmer估计了R中的多级模型。我的虚拟变量之一和连续变量之间存在显着的相互作用。虚拟变量实际上代表一个三级因子。有没有办法我可以直观地绘制此交互关系，以便表示原始变量的所有三个级别？
model <- lmer (
dPrime ~ 1 +
targetRaceDummy2Black + targetRaceDummy2Asian +
targetRaceSelfStatusLLC +
pRaceDummy2Asian + pRaceDummy2Black +
targetRaceDummy2Black*pRaceDummy2Asian +
targetRaceDummy2Asian*pRaceDummy2Asian +
targetRaceDummy2Black*pRaceDummy2Black +
targetRaceDummy2Asian*pRaceDummy2Black +
stigConscIndexLLC +
grpIdIndexLLC +
targetRaceContactLLC +
sdoIndexLLC +
stigConscIndexLLC*pRaceDummy2Asian + stigConscIndexLLC*pRaceDummy2Black
+
sdoIndexLLC*targetRaceDummy2Asian + sdoIndexLLC*targetRaceDummy2Black +
(1 + targetRaceDummy2Black +
targetRaceDummy2Asian +
targetRaceSelfStatusLLC +
targetRaceContactLLC || ResponseID),
data = analysisDataSubset,
REML = FALSE,
na.action = na.omit)
固定效果的摘要如下：
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 6.294e-01 1.433e-01 3.006e+02 4.390 1.57e-05 ***
targetRaceDummy2Black1 -2.329e-01 7.243e-02 2.803e+02 -3.215 0.00146 **
targetRaceDummy2Asian1 -2.294e-01 7.324e-02 3.545e+02 -3.132 0.00188 **
targetRaceSelfStatusLLC 1.655e-03 7.170e-03 5.291e+02 0.231 0.81749
pRaceDummy2Asian 2.755e-01 1.531e-01 2.528e+02 1.799 0.07315 .
pRaceDummy2Black -4.547e-02 1.637e-01 2.412e+02 -0.278 0.78146
stigConscIndexLLC 7.894e-02 3.580e-02 1.944e+02 2.205 0.02861 *
grpIdIndexLLC -3.957e-02 1.789e-02 2.025e+02 -2.212 0.02811 *
targetRaceContactLLC 1.568e-03 1.017e-02 3.909e+02 0.154 0.87761
sdoIndexLLC -7.264e-02 2.347e-02 2.995e+02 -3.095 0.00215 **
targetRaceDummy2Black1:pRaceDummy2Asian -8.389e-02 7.551e-02 2.232e+02 -1.111 0.26777
targetRaceDummy2Asian1:pRaceDummy2Asian 1.354e-01 8.429e-02 3.369e+02 1.606 0.10916
targetRaceDummy2Black1:pRaceDummy2Black 1.972e-01 8.797e-02 2.672e+02 2.242 0.02579 *
targetRaceDummy2Asian1:pRaceDummy2Black 5.433e-02 7.930e-02 3.105e+02 0.685 0.49377
pRaceDummy2Asian:stigConscIndexLLC -1.053e-01 4.869e-02 1.979e+02 -2.162 0.03179 *
pRaceDummy2Black:stigConscIndexLLC -4.983e-04 4.838e-02 2.014e+02 -0.010 0.99179
targetRaceDummy2Asian1:sdoIndexLLC 3.931e-02 2.241e-02 2.969e+02 1.754 0.08044 .
targetRaceDummy2Black1:sdoIndexLLC 2.826e-02 2.178e-02 2.183e+02 1.298 0.19578
我可以使用以下代码成功绘制pRaceDummy2Asian：stigConscIndexLLC交互：
ef1 <- effect(term="pRaceDummy2Asian*stigConscIndexLLC", xlevels= list(pRaceDummy2Asian=c(0, 1)), mod=addControlsInterX)
ef1
efdata1<-as.data.frame(ef1) #convert the effects list to a data frame
efdata1 #print effects data frame
efdata1$pRaceDummy2Asian<-as.factor(efdata1$pRaceDummy2Asian)
ggplot(efdata1, aes(x=stigConscIndexLLC, y=fit, color=pRaceDummy2Asian,group=pRaceDummy2Asian)) +
geom_point() +
geom_line(size=1.2) +
geom_ribbon(aes(ymin=fit-se, ymax=fit+se, fill=pRaceDummy2Asian),alpha=0.3) +
labs(title = "Title", x= "Stigma consciousness", y="Recognition", color="0=white/black 1=asian", fill="0=white/black 1=asian") + theme_classic() + theme(text=element_text(size=20))
有没有办法绘制出显示所有三个种族水平的模型？
