diff --git a/slides/lead_lmm.pdf b/slides/lead_lmm.pdf
index 3ae2517..130dd78 100644
Binary files a/slides/lead_lmm.pdf and b/slides/lead_lmm.pdf differ
diff --git a/slides/lead_lmm.tex b/slides/lead_lmm.tex
index d4c96f9..38857ab 100644
--- a/slides/lead_lmm.tex
+++ b/slides/lead_lmm.tex
@@ -365,6 +365,15 @@
 \section{Hierarchical modeling}
 
 \begin{frame}{HSB data set}
+  \begin{itemize}
+    \item Two levels
+    \begin{itemize}
+      \item Level 1: Student attributes
+      \item Level 2: School attributes
+    \end{itemize}
+  \end{itemize}
+  \vfill
+
   \centering
   \begin{tabular}{llp{10cm}}
     \hline
@@ -474,8 +483,8 @@ $
 \begin{frame}{Results}
   {Random effects}
   \begin{itemize}[<+->]
-    \item The estimate $\hat\sigma^2_{\upsilon_0} = 2.32$ of the variance of
-      mean school performance provides room for improving prediction by
+    \item The (ML) estimate $\hat\sigma^2_{\upsilon_0} = 2.32$ of the variance
+      of mean school performance provides room for improving prediction by
       including additional predictors
     \item However, there is virtually no variation in the dependence of math
       achievement on \texttt{cses} across schools ($\hat\sigma^2_{\upsilon_1} =