Course Description

Many regulations compare a mean or percentile of observed data to a standard or guideline.  If the mean exceeds the standard, the soil/water being sampled is out of compliance.  If more than an allowed percentage (say 10%) of observations exceed the standard, the soil/water is out of compliance.  Such comparisons lead to elevated false positives, false statements of non-compliance.  The statistical approach compares a confidence limit on the mean or percentile rather than the parameter itself to the standard value.  Why should a confidence limit be employed?  What are the drawbacks to using the observed mean or percentile?  After answering these questions, the method for comparing a confidence limit to a standard is illustrated.

Ph.D.

Dennis Helsel

Dr. Dennis Helsel (PhD, Environmental Science and Engineering) has applied statistics to practical issues in environmental sciences for over 36 years. He is the principal of Practical Stats, an environmental statistics training and consulting firm. He has conducted courses in China, Germany, the Netherlands, Finland, Spain, Singapore, and throughout North America.His courses on environmental statistics are currently emerging online at http://practicalstats.com/training .He has authored two textbooks, “Statistical Methods in Water Resources" (2002, with a second edition coming in 2017), and "Statistical Methods for Censored Environmental Data using Minitab and R” (2012, previously titled Nondetects And Data Analysis), as well as 50+ journal articles. Dr. Helsel was the 2003 recipient of the Distinguished Achievement Award from the American Statistical Association’s section on Statistics and the Environment for his training of scientists in applied, practical statistics.

Course curriculum

  • 1

    Comparing Data to Standards Using Statistics: Why and How

    • Comparing Data to Standards Using Statistics: Why and How

    • Discussion