> Less Than Obvious: Statistical Methods for Nondetects
> taught by Dennis Helsel, Ph.D.
> Colorado School of Mines, Golden CO
> August 18-19, 2004
> www.practicalstats.com
>
> For most environmental professionals, the way to deal with
> nondetects is less than obvious. Values below detection or
> reporting limits result from measuring trace amounts of a variety of
> organic and inorganic chemicals. This 2-day course examines methods
> for
> interpreting nondetect data translated from disciplines such as
> medical
> statistics and econometrics. Methods are surveyed for computing
> summary statistics, hypothesis tests, and regression for one or more
> detection limits. Example problems are worked in class, so students
> can confidently take these methods back to their office. The course
> assumes a knowledge of basic statistics, including some familiarity
> with t-tests, linear regression, and simple nonparametric tests like
> the rank-sum test.
> Topics covered in the class include:
> * Why substituting arbitrary values such as one-half the detection
> limit produces unacceptable results.
> * Computing UCL95s and other confidence intervals for data with
> nondetects.
> * Cohen's method -- old technology. What's better now.
> * Parametric and nonparametric tests for censored data.
> * Correlation coefficients with single and multiple detection limits.
> * How to build three types of regression models.
> * Plotting data with nondetects.
> Course content follows the textbook "Nondetects And Data Analysis" by
> Dennis Helsel to be published by Wiley in Summer, 2004. For more
> information on course content and to register for the course, go to
> http://www.practicalstats.com/Pages/lto.html
>
>