\documentclass[11pt,a4rpaper]{article}
\usepackage[top=1.1in,bottom=1in,left=0.9in,headheight=14pt,right=0.9in]{geometry}
\usepackage{paralist,mdwlist,Sweave,amsmath,amsbsy,graphicx,alltt,fancyhdr,url}
\SweaveOpts{engine=R,eps=FALSE,width=10,height=6.5,strip.white=all}
\SweaveOpts{prefix=TRUE,prefix.string=figs/hw9,include=TRUE}
\setkeys{Gin}{width=\textwidth}
\DefineVerbatimEnvironment{Sinput}{Verbatim}
{formatcom={\vspace{-1.5ex}},fontshape=sl,
  fontfamily=courier,fontseries=b, fontsize=\small}
\DefineVerbatimEnvironment{Soutput}{Verbatim}
{formatcom={\vspace{-2.5ex}},fontfamily=courier,fontseries=b,fontsize=\small}
\usepackage{fancyhdr}
\lhead{\sf Mixed-effects models workshop}\chead{\sf Exercises \#1}
\rhead{\sf 2009-07-01 (p. \thepage)}
\lfoot{\sf Loading data from a file and examining it}\cfoot{}\rfoot{}
\pagestyle{fancy}
\newcommand{\code}[1]{\texttt{\small #1}}
\newcommand{\R}{\textsf{R}}
<<initial,echo=FALSE,print=FALSE>>=
options(width=80, show.signif.stars = FALSE)
library(lattice)
lattice.options(default.theme = standard.theme(color = FALSE))
set.seed(123454321)
@ 
\begin{document}
These exercises use data from a study of mathematics skill levels in
students within classrooms within schools.   The data are available in
a comma-separated value (csv) file as
\begin{center}
  \url{http://www-personal.umich.edu/~bwest/classroom.csv}
\end{center}

\begin{enumerate}
\item Start \R{} and create a data frame called \code{classroom} using
  the \code{read.csv} function with the (quoted) file name as the
  first and only argument.  Notice that you can do this in two ways:
  either read directly from the web site by giving the quoted URL as
  the argument or first download the file then enter the file name.
  Try it both ways.  Remember that \code{file.choose()} brings up a
  chooser panel to help you navigate to the file name.
<<echo=FALSE,results=hide>>=
classroom <- read.csv("http://www-personal.umich.edu/~bwest/classroom.csv")
@ 

<<strclass1,echo=FALSE,results=hide>>=
foo <- capture.output(str(classroom))
@ 
\item Check the structure of the data frame using
  \code{str(classroom)}.  Are any of the variables in the data frame
  stored as factors?  Should any of these variables be stored as
  factors?  The first few lines should look like
<<strclass,eval=FALSE>>=
str(classroom)
@   
<<strclass2,echo=FALSE>>=
cat(paste(foo[1:5],collapse="\n"),"\n")
@ 
<<sexchange,echo=FALSE>>=
classroom <- within(classroom,
                {
                    sex <- factor(sex, labels = c("M","F"))
                    minority <- factor(minority, labels = c("N","Y"))
                    classid <- factor(classid)
                    schoolid <- factor(schoolid)
                    childid <- factor(childid)
                })
@ 
\item Check the \code{summary} of this data frame.  Is the summary of
  the \code{sex} variable meaningful?  It happens that the coding for
  \code{sex} is
  \begin{compactdesc}
  \item[0] Male
  \item[1] Female
  \end{compactdesc}
  Convert this variable to a \code{factor} with labels \code{"M"} and
  \code{"F"}.  Check the summary for this variable after conversion.
  You can either ask for a summary of the whole data frame again or
  for a summary of this variable only using
<<sexsummary>>=
summary(classroom$sex)
@ 

\item Convert the \code{minority} variable to a factor with levels
  \code{"N"} and \code{"Y"}.  Convert the \code{childid},
  \code{classid} and \code{schoolid} variables to factors (it is
  easiest to retain the numeric values for the levels).  Check
  \code{str} and the \code{summary} again.  The summary of the factor
  variables should be like
<<subsetsummary>>=
summary(subset(classroom, select = c(sex, minority, childid, classid, schoolid)))
@

\item Save the modified data frame as a file named
  \code{"classroom.rda"}.  Remove the data frame.  Load the file and
  check that the \code{classroom} data frame has the expected structure.
<<saveload,echo=FALSE,results=hide>>=
save(classroom, file = "classroom.rda")
rm(classroom)
exists("classroom")
load("classroom.rda")
@   
<<checkstr>>=
str(classroom)
@ 
\end{enumerate}

If you finish early you could also consider these questions.
\begin{enumerate}
\item I claim that the \code{childid} variable serves no purpose
  because it is the same as the row number.  Check this.
<<checkchildid,echo=FALSE,results=hide>>=
all(row.names(classroom) == classroom$childid)
@   

\item You can remove a variable from a data frame by assigning it the
  special value \code{NULL}.  Try this
<<childid>>=
classroom$childid <- NULL
names(classroom)
@   
\end{enumerate}
\end{document}

%%% Local Variables: 
%%% mode: latex
%%% TeX-master: t
%%% End: