[R-sig-teaching] R-sig-teaching Digest, Vol 59, Issue 9

[Nicholas Horton]

> Date: Tue, 29 Jan 2013 13:38:28 -0700
> From: Allen Cornelius <allencornelius at gmail.com>
> To: "r-sig-teaching at r-project.org" <r-sig-teaching at r-project.org>
> Subject: [R-sig-teaching] Data sets for teaching advanced stats with R
> 
> Knowledgable group,
> 
> I am in the process of designing an advanced statistics class for doctoral level clinical psychology students. Their background in basic statistical concepts is OK, but they have little exposure to statistical software. I am looking for pre existing data sets that they can explore fairly easily, probably using R and R Commander (though some coding would be OK), to illustrate some multivariate analyses (e.g., advanced regression, MANOVA, factor analysis). I don't expect proficiency with the software, but enough exposure to the output of results of analyses to understand what things mean. I am looking for multiple examples of the same analyses, so they get exposure and practice with lots of stuff. 
> 
> Allen
> 

Allen,

You might be interested in checking out the HELP study dataset, which is included as part of the "mosaic" package in R as well as in a variety of formats from the book "SAS and R: Data Management, Statistical Analysis and Graphics" book website at http://www.math.smith.edu/sasr/datasets.php.

This includes deidentified data from a randomized trial of n=470 subjects recruited from a detox center in Boston who were randomized to usual care or an intervention to link them to primary care.  Approximately 200 variables are measured at each of baseline, 6 months, 12 month, 18 month and 24 month follow up.  There are a variety of sociobehavioral, health, and health services questions which can be addressed.

The complete set of instruments can be downloaded from http://www.math.smith.edu/help.

Nick

Nicholas Horton 
Department of Mathematics and Statistics, Smith College
Clark Science Center, Northampton, MA 01063-0001
http://www.math.smith.edu/~nhorton