### $Id: README,v 1.2 1998/04/15 13:16:46 bates Exp $

This directory contains a "data" directory with some of the data sets
from the book "SAS System for Mixed Models" by Littell, Milliken,
Stroup, and Wolfinger (1996) SAS Institute.  These are in the form of
S-PLUS/R source code (.q files) that produce groupedData objects.
Another directory "transcripts" contains sample S transcripts (.St
files) for analyses to parallel those in the Littell et al. book but
using the lme function from the nlme library in S-PLUS.  The
transcripts directory also contains .q files that give only the S-PLUS 
code to produce the transcript.  They can be used with Splus BATCH to
check that your copy of the nlme library is producing the expected
results.

Many of these analyses can also be done in R using the lme library.

The .q files are S source code that should be read with the source
function in S-PLUS or in R.  Alternatively, in S-PLUS under Unix you
can create a library of the data sets with shell commands such as

 mkdir .Data
 cat data/*.q | Splus

There are a few comment lines at the beginning of each data file to
describe the data and which data set in "SAS System for Mixed Models"
it corresponds to.

The .St files are transcripts of S sessions.  They are best viewed
with the ESS (Emacs Speaks Statistics) package for GNU emacs and
Xemacs available at http://franz.stat.wisc.edu/pub/ESS/.  The commands 
to attach the required library (nlme in S-PLUS, lme in R) are not
given in each transcript.  The command to source the relevant data set 
is also omitted in the transcript.

At present the calls to graphics routines will only work in S-PLUS
because they depend on the trellis graphics library.

Please see the paper "lme for SAS PROC MIXED users" in this directory
for some general comments on using lme versus using SAS PROC MIXED.

## Local Variables:
## mode:indented-text
## End:
