[R] Do you use R for data manipulation?

Mike Lawrence Mike.Lawrence at dal.ca
Wed May 6 16:21:33 CEST 2009


I work in cognitive science where we collect one or more data files
per participant in an experiment then merge those files to perform
subsequent analyses. Sometimes some files are in wide format and
others are in long format, necessitating reshaping. I've found R
entirely satisfactory for this.*

Indeed, I would be wary of an approach that attempts data manipulation
*outside* of R as I'm of the "raw data in, results out" school of
thought that it's dangerous to isolate your data manipulation from
your record of analysis. If you leave your raw data files untouched
and perform all manipulation & analysis in one system (R) then there
is a complete record of what's happened to the data from start to
finish and it's easier to catch/correct errors.

The reshape package is great for reshaping between long & wide data
formats, and the ply package is great for computing summary statistics
within cells of the design.

Mike

*note: I typically use Python for data collection (showing visual
stimuli, recording responses, etc), but have it spit out raw text
files of the trial-by-trial data, and thus use it for only a bare
minimum of processing.

--
Mike Lawrence
Graduate Student
Department of Psychology
Dalhousie University

Looking to arrange a meeting? Check my public calendar:
http://tr.im/mikes_public_calendar

~ Certainty is folly... I think. ~




More information about the R-help mailing list