Summary of the maximum likelihood mixed-effects model analysis reporting fixed- and random-effects separately.

The fixed effects are reported following the conventions of least squares ANOVA (e.g. Table 2). The R lmer function currently does not give P values due to the difficulties of calculating them for mixed-effects models. The random effects section reports the super-population variance components on the variance and standard deviation scales (the SD is simply the square root of the variance component) with likelihood ratio tests of the change in deviance on removing each random effect from the model in turn. Each variance component is expressed as a percentage of the summed total for all 6 six random effects (lower column 3: ‘%’).