Modeling Posidonia oceanica growth data: from linear to generalized linear mixed models

• DOI: 10.1002/env.1063
• Publication Date: 2010-12-29T11:41:31Z
• Authors: G. Lovison; M. Sciandra; A. Tomasello; S. Calvo
• Keywords: Generalized additive model; Mixed model; Posidonia oceanica; Data set

The statistical analysis of annual growth Posidonia oceanica is traditionally carried out through Gaussian linear models applied to untransformed, or log-transformed, data. In this paper, we claim that there are good reasons for re-considering established practice, since real data on often violate the assumptions models, and show class Generalized Linear Models (GLMs) represents a useful alternative handling such violations. By analyzing Sicily PosiData-1, dataset P. gathered in period 2000–2002 along coasts Sicily, find majority cases Normality rejected effect age nonlinear. A GLM with Gamma distribution identity log link appears be satisfactory choice most cases. Furthermore, when back-dating techniques employed, each plant provides longitudinal set dependent data, proper should take dependence into account. We Mixed (GLMM), an extension GLM's, effective way analyze Again, by using examples taken from misleading results can obtained if ignored other techniques, like sub-sampling, not option overcoming so-called "pseudo-replications" problem. Copyright © 2010 John Wiley & Sons, Ltd.