diff --git a/vignettes/pisa2012.Rmd b/vignettes/pisa2012.Rmd
index 07a0172..a82c4e0 100644
--- a/vignettes/pisa2012.Rmd
+++ b/vignettes/pisa2012.Rmd
@@ -76,7 +76,7 @@ It seems that there exist some differences among countries included in PISA. Let
 # Factor Merger
 
 ```{r}
-pisaIdxSubset <- sample(1:nrow(pisa2012), size = 5000)
+pisaIdxSubset <- sample(1:nrow(pisa2012), size = 500)
 pisaFM <- mergeFactors(pisa2012[pisaIdxSubset, 1:3],
                        factor(pisa2012$CNT[pisaIdxSubset]))
 
@@ -84,7 +84,7 @@ pisaFM
 plot(pisaFM, responsePanel = "profile")
 ```
 It's faster to use 'hclust' method on a big dataset.
-```{r}
+```{r, eval = FALSE}
 pisaFMHClustMath <- mergeFactors(pisa2012[, 1:3],
                        factor(pisa2012$CNT),
                        method = "hclust",
@@ -101,7 +101,7 @@ plot(pisaFMHClust)
 ```
 Let's now have a try using European countries only.
 
-```{r}
+```{r, eval = FALSE}
 
 pisaEuropean <- filter(pisa2012, CNT %in% c("Austria", "Belgium", "Bulgaria",
                                             "Czech Republic", "Germany", "Denmark",
@@ -124,7 +124,7 @@ plot(pisaFMHClustEurope)
 
 # Another factor - parent occupation
 
-```{r}
+```{r, eval = FALSE}
 which <- (ocod$code %>% substr(0,2)) %in% (table(ocod$code %>% substr(0,2)) 
                                            %>% as.data.frame() %>% filter(Freq > 1000))$Var1