PHD Discussions Logo

Ask, Learn and Accelerate in your PhD Research

Question Icon Post Your Answer

Question Icon

2 years ago in Ecology , Statistics By Amol

Are there publications that use lsmeans or glht after mixed-effects models (lmer) in ecology?

My statistical reviewer asked me to justify my choice of post-hoc test after a significant lmer model. I know lsmeans (now emmeans) or glht from multcomp are standard, but I need canonical examples from the literature to cite as precedent. I'm looking for papers that clearly show their R code or state their methods in a replicable way, preferably in community or plant ecology.

 

All Answers (1 Answers In All)

By Joshna Answered 2 years ago

Absolutely. For solid, methodologically clear examples, I often point students to papers in Methods in Ecology and Evolution or Ecological Applications. I have seen a particularly well-documented use in Visser et al. (2016, Journal of Ecology), "Climate vs. soil factors in local adaptation of two common plant species," where they detail their lmer models and subsequent lsmeans contrasts in the supplement. Also, check work by Bjornstad (emphasizing glht for temporal series) or any recent paper from the Zuur group. The key is to cite studies that explicitly name the packages (lme4, emmeans, multcomp) and discuss adjustments like Tukey or FDR for multiple comparisons, which shows a deep understanding of the process.

 

Your Answer