After starting R as discussed in the resources/Getting .. started ... Rcmdr
I then entered my data set (Data-> New dataset) and entered my height data
followed by quit.  I called it heights.

Selecting Statistics->Summaries->Active Data set gave the following for my
data:
> summary(heights)
      var1      
 Min.   :56.00  
 1st Qu.:62.25  
 Median :66.00  
 Mean   :67.30  
 3rd Qu.:70.00  
 Max.   :93.00
Note these our different values than mine.  Also than the textbook's example
which does the same as mine.

I then did Graphs->Boxplot which produced a vertical boxplot.

Looking at a different page made me realize I had to enter:
boxplot(heights$var1, xlab="Height (cm)", horizontal=TRUE)
directly in the command window followed by hitting the SUBMIT button.

Finally Graph->Save Graph to File
allowed me to save the graph as a pdf.

UPDATE:
It turns out the quantile function is more flexible in that it allows you to
enter different types of quantiles.  It turns out that type 6 is what we need.
Type "help(quantile)" in R for help.
> quantile(Heights[,1],type=6)
   0%   25%   50%   75%  100% 
56.00 60.50 66.00 70.25 93.00 
and this is consistent with our textbook.  The [,1] indicates we want the
first column of our Heights 2-D matrix.

Unfortunately getting this change to work with our boxplot command is non-trivial.
This page:
http://r.789695.n4.nabble.com/how-to-change-the-quot-quantile-quot-method-in-bwplot-td897865.html
discusses it using the bwplot command from the lattice package.  Do this:
library(lattice)
my.boxplot.stats<-function(y,subscripts,...)boxplot.stats(quantile(y,type=1),...)
bwplot(x=Heights[,1],stats=my.boxplot.stats)
On that page there is the entry

for teaching purposes I wrote a corresponding function; cf.
qbxp.stats (as well as qboxplot ...) in package MKmisc. 

Unfortunately that package requires version 2.11 which is too new.
