Please use this identifier to cite or link to this item:
|Title:||Applications guide for statistical analyses in dredged sediment evaluations|
|Authors:||Dredging Operations Technical Support Program (U.S.)|
Clarke, Joan U.
Brandon, Dennis L.
|Keywords:||Dredged sediment evaluation|
Equality of variances assumption
Less-than detection limit data
Statistical error rates
|Publisher:||Environmental Laboratory (U.S.)|
Engineer Research and Development Center (U.S.)
Abstract: Dredged sediment evaluations often require statistical analysis of chemical or biological test results. However, the resulting data are frequently problematic for standard statistical procedures because of improper experimental design, insufficient replication, failure to meet statistical test assumptions, outliers, and missing or below detection limit observations. Such nonideal data can seriously affect the error rates of statistical tests. This in turn increases the likelihood of drawing false inferences concerning the potential of a dredged sediment for adverse biological effects. This report explores the impact of nonideal data on the performance of statistical tests recommended for dredged sediment evaluations. Procedures are discussed for basic experimental design, testing assumptions and outliers, assessing statistical power, analyzing less-than detection limit data, and interpreting statistical test results.
|Rights:||Approved for public release; distribution is unlimited.|
|Appears in Collections:||Miscellaneous Paper|