Statistical Methods for Environmental Data - NWAEP Lunch & Learn - Portland

"Statistical Methods for Censored Environmental Data"

Abstract:
Low-level concentrations of environmental organic and inorganic chemicals are often reported as "nondetected" or "less-than" values. These censored concentrations are known only to be somewhere between zero and the laboratory's reporting level (RL). Over the past 20 years the fields of medical survival analysis and industrial reliability analysis have greatly improved methods for analyzing censored data--those observations reported only as being above or below a threshold value. Today, methods for  correctly summarizing, testing statistical hypothesis, and modeling cause-and-effect (regression analyses) censored data are included in statistics software.
 Decades-old methods originally applied only to "greater-thans" (right censored data) can also be applied to the "less-thans" (left censored data) of low-level environmental concentrations. Unfortunately, regulators of the environmental community have generally not incorporated these procedures or required their use when permit holders submit results of compliance or baseline monitoring efforts. This talk provides an overview of methods currently available for interpreting data with nondetects.

Date:  Friday, April 19th

Time: 11:30 to Noon - Meet and greet

             Noon to 1pm - Talk followed by Q&A

Location: Central Portland Library, US Bank Room

                    801 SW 10th Ave, Portland 97205

Please contact Makary Hutson at 509.981.7542 or makaryh@gmail.com with any questions.