5

I am fitting several models and I am plotting them with sjPlot package using plot_models function. I want just to plot some of the coefficients and also reordering them to make easier to interpret. However, I have troubles in the second issue.

I have tried the order.terms but is just for plot_model, the auto.label=FALSE but just remove all the labels...

fit1<- glm(y ~ x1*var1 + var2 + var3, data=mydata, family = poisson)
fit2<- glm(y ~ x2*var1 + var2 + var3, data=CARMA_all, family = poisson) 
fit3<- glm(y ~ x3*var1 + var2 + var3, data=CARMA_all, family = poisson) 

plot_models(fit1,fit2,fit3, m.labels = c("x1", "x2", "x3"),show.values =T,digits=3, show.p = FALSE, p.shape = TRUE, rm.terms = c("var1,"var2","var3"),vline.color="grey",grid.breaks=c(0.95,0.96,0.97,0.98,0.99,1,1.01,1.02,1.03,1.04,1.05), colors = "Dark2")

When I plot this I have in the plot first the estimate var1:x3, then x3, then var1:x2, x2, var1:x1, x1. I need to reorder to have x1, var1:x1, x2, var1:x2, x3, var1:x3

Sara
  • 185
  • 1
  • 9
  • Have you tried the `order.terms` argument? See [examples here](https://strengejacke.github.io/sjPlot/articles/plot_model_estimates.html) – Daniel Jul 30 '19 at 20:50
  • @Daniel yes, it is indeed in the description. Thank you! – Sara Jul 31 '19 at 08:51

0 Answers0