Statistical properties of relative weight distributions of four salmonid species and their sampling implications

North American Journal of Fisheries Management
By:  and 

Links

Abstract

We assessed relative weight (Wr) distributions among 291 samples of stock-to-quality-length brook trout Salvelinus fontinalis, brown trout Salmo trutta, rainbow trout Oncorhynchus mykiss, and cutthroat trout O. clarki from lentic and lotic habitats. Statistics describing Wr sample distributions varied slightly among species and habitat types. The average sample was leptokurtotic and slightly skewed to the right with a standard deviation of about 10, but the shapes of Wr distributions varied widely among samples. Twenty-two percent of the samples had nonnormal distributions, suggesting the need to evaluate sample distributions before applying statistical tests to determine whether assumptions are met. In general, our findings indicate that samples of about 100 stock-to-quality-length fish are needed to obtain confidence interval widths of four Wr units around the mean. Power analysis revealed that samples of about 50 stock-to-quality-length fish are needed to detect a 2% change in mean Wr at a relatively high level of power (beta = 0.01, alpha = 0.05).
Publication type Article
Publication Subtype Journal Article
Title Statistical properties of relative weight distributions of four salmonid species and their sampling implications
Series title North American Journal of Fisheries Management
DOI 10.1577/1548-8675(2001)021<0666:SPORWD>2.0.CO;2
Volume 21
Issue 3
Year Published 2001
Language English
Larger Work Type Article
Larger Work Subtype Journal Article
Larger Work Title North American Journal of Fisheries Management
First page 666
Last page 670
Google Analytic Metrics Metrics page
Additional publication details