부산대학교 도서관

This article discusses the issue of data variation in comparative biological analyses, specifically focusing on metabolic rate data. The authors argue against the common practice of removing statistical outliers from regression analyses, providing five reasons to support their decision. They use the example of a biphasic analysis of the ontogenetic metabolic scaling of the American eel to illustrate their points. The authors conclude that their analysis, which includes all data points, better represents the variation in the metabolic data and supports their conclusion of biphasic metabolic scaling in the eel species. [Extracted from the article]