most conservative df for mixed effects anova
Dear Carrie, The most conservative number IMHO is the sum of the number fixed effects parameters and the number of random effects parameters (in case of a random intercept: 1 level = 1 parameter). Het most liberate number would replace the number random effects parameters with the number of random effect hyperparameters (a random intercept = 1 variance = 1 hyperparameter). Best regards, ir. Thierry Onkelinx Statisticus / Statistician Vlaamse Overheid / Government of Flanders INSTITUUT VOOR NATUUR- EN BOSONDERZOEK / RESEARCH INSTITUTE FOR NATURE AND FOREST Team Biometrie & Kwaliteitszorg / Team Biometrics & Quality Assurance thierry.onkelinx at inbo.be Havenlaan 88 bus 73, 1000 Brussel www.inbo.be /////////////////////////////////////////////////////////////////////////////////////////// To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher The plural of anecdote is not data. ~ Roger Brinner The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. ~ John Tukey /////////////////////////////////////////////////////////////////////////////////////////// <https://www.inbo.be> Op di 15 okt. 2019 om 15:20 schreef Carrie Perkins <cperk at terpmail.umd.edu>:
Hello!
I have data from an experiment and would like to run an anova with fixed
and random effects in R. Here is information on the data:
In the experiment, 3 replicates of 48 plant genotypes were planted into
each of 4 salinity treatments. This resulted in a total of 144 individuals
per treatment, amounting to a grand total of 576 individuals in the whole
experiment. Within each treatment, random sets of 24 plants were grouped
into a total of 6 pools to make it easier to monitor salinity levels. I
would like to model these pools as random Experimental Units.
I would like to make Experimental Unit the random effect and look at the
treatment X genotype interaction as fixed effects.
lmer_model_3 <- aov(Y~Genotype*Treatment + Error(1|Experimental Unit),
data=dataframe)
What would be the most conservative method for calculating degrees of
freedom for the random effects term of an anova? When I've tried
researching this question online, I find a lot of information on
calculating degrees of freedom for basic 1- and 2-way anovas (which I
understand) but I can't find clear guidance on how to calculate the degrees
of freedom for anovas with random effects.
Thank you!
Sincerely,
Carrie Perkins
[[alternative HTML version deleted]]
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models