[R-sig-ME] Report results for GLMM
Rafael Lima Oliveira
o||ve|r@r|@ue@b @end|ng |rom gm@||@com
Thu Nov 17 21:49:06 CET 2022
Dear list members,
I'm using gamma-based GLMMs to estimate the effect of a set of continuous
environmental variables (salinity, temperature, pH, dissolved oxygen,
turbidity and depth) on rarefied species richness and fish density along
four coastal environments.
I´m using the “glmer” function from the “lme4” package and the link
function used was the logarithmic. For each locality I ran a full model and
after that, I applied a model selection using the dredge function from the
MuMIn package. After performing this procedure, I selected five models
ranked by the smallest AIC, and its related delta (ΔAIC) and weight.
I'm looking at other papers in my field to find examples of what I
should report
in my results section.
In my results I report two tables, one for each response variable with the
information from the best model (smallest AIC) from the dredge output and Anova
output (Analysis of Deviance table - Type II Wald chisquare test). Besides
that, I report effect plots only for each statistically significant
predictor variable. Should I report Pseudo-R-squared for the best model?
Instead of reporting Anova table should I report a table with the five
models ranked by smallest AIC from the dredge output?
Thanks for any answer!
Rafael
--
*Rafael Lima Oliveira*
Doutorando em Biologia Animal
Universidade Federal do Espírito Santo - UFES
Laboratório de Ecologia de peixes marinhos - CEUNES/UFES
*Contato:* (75) 98873-1548 / (27) 99526-3612
*E-mail alternativo:* rafael.l.oliveira using edu.ufes.br
*Currículo Lattes*: http://lattes.cnpq.br/5215941704013482
>)))º> >)))º> >)))º> >)))º>
[[alternative HTML version deleted]]
More information about the R-sig-mixed-models
mailing list