Generalized linear mixed models (GLMM) in community ecology studies using the R statistical environment

Data analysis in community ecology often has certain difficulties, since standard parametric methods are inapplicable due to the fact that ecological data rarely normal distributed, there are no linear relationships between variables, there may be collinearity between explanatory variables and overdispersion in data sets. The proposed article considers an approach based on the use of regression generalized linear mixed models (GLMM), which allows analyzing data from synecological studies taking into account the above difficulties, as well as including not only quantitative but also qualitative predictors in the analysis. Using our own results of studies of ground beetle assemblages in several types of forests, the stages of GLMM implementation in the R statistical environment are demonstrated. A simple program code for GLMM is proposed, available for use by novice researchers. The protocol for exploratory data analysis is considered in detail, the use of appropriate packages, including lme4, performance, car, etc. is justified.