NADA for R
R is a free and freely-available implementation of the S-language for statistical computing originally developed by John Chambers and others at Bell Labs. It has a Unix-like interface, but compiled versions run under the Unix, Windows and Macintosh operating systems. It is a powerful environment for statistical computing -- newly-developed methods are frequently released with an R implementation. R operates under the GNU license and contributed packages are developed by volunteers worldwide.
NADA for R is a user-contributed package available from the CRAN (Comprehensive R Archive Network) site. See the Packages link to the left and look for "NADA". It was written by Lopaka Lee of the US Geological Survey, and implements the procedures found in the textbook Statistics for Censored Environmental Data using Minitab and R by Helsel (2012), published by Wiley. Methods are included for computing descriptive statistics, hypothesis tests, correlation and regression for left-censored (nondetect) data. NADA for R can be downloaded and installed from within R using the install.packages command.
USER GUIDE: A user's guide to NADA for R is available on our NADA downloads page. The guide is the handout from our August 2006 NADA for R workshop for the American Statistical Association. Through example exercises, the use of the NADA for R package is demonstrated. Datasets used in the exercises come with the NADA for R package, so you can perform each exercise yourself to become familiar with all the NADA for R parametric and nonparametric functions.
USCORES FUNCTION: A uscores function for censored data is now available on the NADA downloads page. This is not included within the NADA for R package.