2.6. Data analysis
The proportion of roots colonized by the various fungal treatments
from the two screening experiments and the additional
experiments were analyzed using logistic regression with random
effects using the R package lme4 (Bates et al., 2015), with fungal
treatment, sampling date, root part, and other endophyte presence
as fixed effects and block nested in experiment as random effects.
The control plant data was not included in these analyses. A means
comparison, using the glht function in the R package multcomp
(Hothorn et al., 2008) was used to separate the differences in proportional
colonization among treatments. This method adjusts pvalues
for the number of comparisons made. A similar analysis
was done using other endophyte presence as the dependent
variable.