Political scientist Drew Conway (he of the awesome live-tweeting social analysis demo) lists his top 10 must-have packages for social scientists using R. They are:
- Zelig, for a uniform regression methodology
- ggplot2, for graphics
- Statnet and igraph, for social network analysis
- plyr, for data manipulation
- Amelia II, for missing data imputation
- nlme, for nonlinear mixed-effects models
- SNOW and Rmpi for parallel computing (what, no love for foreach?)
- xtable/apsrtable for print-ready tables
- plm, for analysis of panel data, and
- sqldfm, for using SQL to process R data
Check out the full article for Drew's discussion of each package and why it makes the top 10 for social scientists.
Zero Intelligence Agents: Must-Have R Packages for Social Scientists
What our social scientists want is two packages, namely:
require(spss)
and:
require(spoonfeedingandclickyinterfaceplease)
But seriously (what, me serious?) are there R packages to do what AMOS for SPSS does? And that do Multi-level modelling as SAS 9.2 does? I say SAS 9.2 because we had a request from a user for 9.2 because 9.1 didn't have enough levels or something... Anyway, those are the things our social scientists seem to use. I know nothing about them. I must be doing anti-social science. Or 'science' as we call it. I knew I couldn't be serious for long.
Posted by: Barry | December 18, 2009 at 09:25
@Barry- See lmer,nlme, and Winbugs4r (and of course, win bugs) for multilevel modelling (also see the Multi-level modeling book by Gelman and Hill). RE: Amos, there are several R packages for structural equation modelling, such search the cran site.
Posted by: Frank | December 18, 2009 at 17:27
Barry: For SEM, take a look at OpenMX:
http://openmx.psyc.virginia.edu/wiki/main-page
Posted by: Ali | December 19, 2009 at 14:29