diff --git a/build/make_example.r b/build/make_example.r
deleted file mode 100644
index 5815b05..0000000
--- a/build/make_example.r
+++ /dev/null
@@ -1,31 +0,0 @@
-# reproducible:
-set.seed(123)
-
-# cts and categorical predictors:
-x <- rep(1:10, times=3)
-g <- as.factor(rep(c('a', 'b', 'c'), each=10))
-
-# parameters
-a <- 10
-b <- -0.25
-s <- 1
-r <- 3
-
-# random effects:
-re <- rnorm(nlevels(g), 0, r)
-names(re) <- levels(g)
-
-# linear response
-y <- a + b*x + re[g] + rnorm(length(x), 0, s)
-
-# poisson response:
-lambda <- exp((a + b*x + re[g])/10)
-z <- rpois(length(x), lambda)
-
-
-
-
-
-example <- data.frame(y=y, x=x, g=g, z=z)
-save(example, file='data/example.rda')
-
diff --git a/build/make_simdata.r b/build/make_simdata.r
index 34ad914..78fa94c 100644
--- a/build/make_simdata.r
+++ b/build/make_simdata.r
@@ -27,5 +27,5 @@ z <- rpois(length(x), lambda)
 
 
 simdata <- data.frame(y=y, x=x, g=g, z=z)
-save(example, file='data/simdata.rda')
+save(simdata, file='data/simdata.rda')
 
diff --git a/data/example.rda b/data/example.rda
deleted file mode 100644
index a4d11b8..0000000
Binary files a/data/example.rda and /dev/null differ
diff --git a/data/simdata.rda b/data/simdata.rda
index 2f4a785..6203719 100644
Binary files a/data/simdata.rda and b/data/simdata.rda differ
diff --git a/vignettes/examples.Rmd b/vignettes/examples.Rmd
index 64d7cd9..a50f553 100644
--- a/vignettes/examples.Rmd
+++ b/vignettes/examples.Rmd
@@ -13,11 +13,11 @@ simrOptions(nsim=25, progress=FALSE, observedPowerWarning=FALSE)
 
 This vignette provides examples of some of the hypothesis tests that can be specified in `simr`. The function `doTest` can be used to apply a test to an input model, which lets you check that the test works before running a power simulation.
 
-Documentation for the test specification functions can be found in the online help at `?tests`.
+Documentation for the test specification functions can be found in the help system at `?tests`.
 
 ## Binomial GLMM with a categorical predictor
 
-The first example comes from the online help for `glmer`. The data frame `cbpp` contains data on contagious bovine pleuropneumonia. An observation variable is added to allow for overdispersion. Note that the response is specified using `cbind` --- `simr` expects a binomial model to be in this form.
+The first example comes from the help page for `glmer`. The data frame `cbpp` contains data on contagious bovine pleuropneumonia. An observation variable is added to allow for overdispersion. Note that the response is specified using `cbind` --- `simr` expects a binomial model to be in this form.
 
 ```{r}
 cbpp$obs <- 1:nrow(cbpp)