Summer 2010: Intro
Home | General: Data Structure -> |
Info
This is the website for the Summer 2010 meetings of the UPenn R Study Group. It has been organized by Josef Fruehwald and attended mostly by researchers affiliated with the Linguistics and Psychology Departments at Penn.
Check out the Google Group for e-mail updates:
http://groups.google.com/group/penn-r-study-2010
Agenda for the R Study Group 2010
Hopefully, this Study Group will operate more like a workshop. That is, it should be maximally flexible and participatory. It would be most useful if we brought datasets which we care about to work with.
Goals for Data Structure and Manipulation
The first half of the workshop will devoted to structuring, manipulating and aggregating your data. Time spent figuring out how to do these things on your own for the first time constitutes an enormous time suck. I think that improving our abilities in these areas will vastly improve our data analysis efficiency.
- General: Discuss the ideal and most flexible formats for data.
- General: Discuss data collection and storage.
- R: Reshaping Data. (
reshape package) - R: Manipulating And Subsetting Data.
- R: Data Aggregation. (
reshape andplyr packages) - R: Efficient workflows.
Goals for Graphical Display
The second half of the workshop will be devoted to generating appropriate and effective graphics,
focusing especially on using the
- General: Discuss appropriate kinds and designs of graphics.
- R: Designing graphics in
ggplot2 - R: Fine tuning graphics in
ggplot2
Sources and References
Data Manipulation
- Data Manipulation with R, Phil Spector
This is a pretty good, and broad book on how to do data manipulation - Plyr webpage
A plyr course
Plyr documentation[pdf]
Plyr is an excellent data aggregation package, and also allows for powerful batch application of a function over subsets of your data. - Reshape webpage
Intro to Reshape
Reshape is a package for reshaping your data.
Graphical Display
- The Visual Display of Quantitiative Information, Edward Tufte
A classic. Sometimes called the Strunk and White of data visualization, but they always mean it as a compliment. - Junk Charts
A blog on what not to do - Eager Eyes:Blog
An Infoviz Blog - ggplot2 website