Classification of the coefficient of variation in experiments with Macrobrachium amazônicum

Abstract

The objective of this work is to propose classification ranges of the Coefficients of Variation (CV's) of the biometric variables frequently studied in experiments with Amazonian shrimp species Macrobrachium amazonicum (HELLER, 1862). CV's were collected in several experimental studies with the species published between 2005 and 2016 for the variables: final weight (PF), survival (SOB), productivity (PROD), total length (CT), feed conversion (CA) and gain of weight (GP). Kolmogorov-Smirnov, modified by LILLIEFORS (1967) was used to test the hypothesis of normality at 5%. The CV data that met normality were used to prepare the classification ranges based on the mean (X ̅) and standard deviation (S). For variables with unmet normality, classification ranges were defined based on the interquartile median (Md) and pseudo-sigma (PS). The coefficients of variation in studies with Amazonian shrimp have peculiar characteristics and differ from the ranges proposed by Gomes (1990), making them inappropriate for most of the variables analyzed. Therefore, with the assumption of normality of the data met, the proposed CV classification ranges based on the two methodologies evaluated are more robust, allowing more consistent data analysis and greater precision in decision making. In this sense, it is suggested to use the classification ranges of the coefficients of variation found in this work to determine the experimental precision in works involving the species.