Software - Resident software

The Uncensor program estimates the mean, standard deviation, variance, and the confidence interval on the mean of left-censored data sets. At present, the program can provide estimates for normal and two-parameter lognormal data sets. Large (N>20) and small (N<20) data sets can be handled. To download a zipped copy of Version 4.0 of the software, click on the link below. You will need the shareware program PKUNZIP.EXE from PKWARE Inc. to expand this file. To download the shareware, click on the link in the previous sentence to reach PKWARE's web site. The manual is also available in PDF format by using the Adobe Acrobat Reader. The reader can be downloaded for free, also by accessing the Adobe link in the previous sentence.

Bootstrap and Bootstrap Users Manual

BOOTSTRAP is designed to calculate a community NOEC (no observed effect concentration) using a bootstrap method. Starting with NOEC values for a set of representative species with respect to a certain toxicant, assume that the available data set is a random sample from some larger population of NOEC values. Another population with the same parameters as the unknown one can be built up by taking a large number of samples from the available data set. To resample a data set, one takes repeated samples ("resamples") with replacement and calculates parameter estimates for each sample. The mean and standard deviation of these estimates are then obtained.

Samplesize and Samplesize Users Manual

This program calculates N, the minimum number of samples (or replicates) necessary to obtain a desired confidence level. The methods are outlined in Quantitative Methods in Aquatic Ecology by M.C. Newman pp 42-45.

Resample

This program estimates the HC5 and its 95% confidence interval from species senstivity data using a bootstrap method. The bootstrap approach has the advantage over parametric methods because it requires no assumption of a particular distribution for the data. The FORTRAN code was developed by Tyler Christensen.