diff --git a/07-the-model-workflow.Rmd b/07-the-model-workflow.Rmd
index 3ef47ba..99395ef 100644
--- a/07-the-model-workflow.Rmd
+++ b/07-the-model-workflow.Rmd
@@ -183,6 +183,8 @@ A number of multilevel models have standardized on a formula specification devis
 
 ```r
 library(lme4)
+data(Orthodont, package = "nlme") # load the Orthodont data from the nlme package
+
 lmer(distance ~ Sex + (age | Subject), data = Orthodont)
 ```
 
@@ -190,10 +192,6 @@ The effect of this is that each subject will have an estimated intercept and slo
 
 The problem is that standard R methods can't properly process this formula: 
 
-```{r echo=FALSE}
-data(Orthodont, package = "nlme")
-```
-
 ```{r workflows-rand-mm, error=TRUE}
 model.matrix(distance ~ Sex + (age | Subject), data = Orthodont)
 ```