Despite its usefulness, R-squared has limitations. It does not indicate whether a regression model is appropriate; it merely describes the proportion of variance explained by the model. Moreover, a high R-squared value does not necessarily mean that the model is good; it could be due to overfitting, especially if the model includes many independent variables. In toxicology, where biological systems are complex and multifactorial, relying solely on R-squared without considering other statistical diagnostics can lead to misleading conclusions.
