my_linear_mixed_models_script.R

library(tidyverse)
library(lme4)
library(ggeffects)

load("dragons.RData")
head(dragons)

basic_lm <- lm(testScore ~ bodyLength2, data = dragons)
summary(basic_lm)

dragons$bodyLength2 <- scale(dragons$bodyLength, center = T, scale = T)


ggplot(dragons, aes(bodyLength2, testScore)) +
  geom_point() +
  geom_smooth(method = "lm")

plot(basic_lm, which = 1)
plot(basic_lm, which = 2)

ggplot(dragons, aes(mountainRange, testScore)) +
  geom_boxplot()

(colour_plot <- ggplot(dragons, aes(bodyLength, testScore, colour = mountainRange)) +
    geom_point(size = 2) +
    theme_classic() +
    theme(legend.position = "right"))

(split_plot <- ggplot(aes(bodyLength, testScore), data = dragons) +
    geom_point() +
    facet_wrap(~ mountainRange) +
    xlab("length") +
    ylab("test score"))

mountain_lm <- lm(testScore ~ bodyLength2 + mountainRange, data = dragons)
summary(mountain_lm)


# Fixed effect -- predictive variable
# Random effect -- colloquially, the grouping factors
# 
# Note: Random effects should have at least five levels. If we had sex as a variable with two levels,
# we would specify it is a fixed, not random effect

mixed_lmer <- lmer(testScore ~ bodyLength2 + (1|mountainRange), data = dragons)
summary(mixed_lmer)

# Nested and (partially) crossed random effects
# 
# Nested random effects, the factor appears ONLY within a particular level of another factor
# Crossed random effects, given factor appears in more than one level of another factor (same dragon, multiple mountains)
# 
# leafLength ~ treatment + (1|Bad/Plant/Leaf) + (1|Season)
#                           ^^^^^^^^^^^^^^^^     ^^^^^^^^
#                           nestedeffects    crossed effects

dragons <- within(dragons, sample <- factor(mountainRange:site)) # turning implicit nesting into explicit one

mixed_lmer2 <- lmer(testScore ~ bodyLength2 + (1|mountainRange) + (1|sample), data = dragons)
summary(mixed_lmer2)

(mm_plot <- ggplot(dragons, aes(x = bodyLength, y = testScore, colour = site)) + 
    facet_wrap(~ mountainRange, nrow = 2) +
    geom_point(alpha = 0.5) +
    theme_classic() +
    geom_line(data = cbind(dragons, pred = predict(mixed_lmer2)), aes(y = pred), linewidth = 0.5) +
    theme(panel.spacing = unit(2, "lines"))
  )

mixed_ranslope <- lmer(testScore ~ bodyLength2 + (1 + bodyLength2|mountainRange/site), data = dragons)
summary(mixed_ranslope)

(mm_plot <- ggplot(dragons, aes(x = bodyLength, y = testScore, colour = site)) + 
    facet_wrap(~ mountainRange, nrow = 2) +
    geom_point(alpha = 0.5) +
    theme_classic() +
    geom_line(data = cbind(dragons, pred = predict(mixed_ranslope)), aes(y = pred), linewidth = 0.5) +
    theme(panel.spacing = unit(2, "lines"))
)

pred_mm <- ggpredict(mixed_lmer2, terms = c("bodyLength2"))

(ggplot(pred_mm) + 
    geom_line(aes(x = x, y = predicted)) +
    geom_ribbon(aes(x = x, ymin = predicted - std.error, ymax = predicted + std.error),
                fill = "lightgrey", alpha = 0.5) +
    geom_point(data = dragons,
               aes(x = bodyLength2, y = testScore, colour = mountainRange)) +
    labs(x = "Body Length (indexed)", y = "Test Score",
         title = "Body length does not affect intelligence in dragons") +
    theme_minimal()
)

full_lmer <- lmer(testScore ~ bodyLength2 + (1|mountainRange) + (1|mountainRange:site), data = dragons, REML = F)
reduced_lmer <- lmer(testScore ~ 1 + (1|mountainRange) + (1|mountainRange:site), data = dragons, REML = F)

anova(reduced_lmer, full_lmer)