###INSTALL PACKAGES###
# uncomment and run if you need to install
# install.packages("lme4")
# install.packages("r2mlm")
# install.packages("dplyr")

###LOAD PACKAGES###
library(lme4)
library(r2mlm)
library(dplyr)

###LOAD DATA###
recall <- read.csv("https://stats.oarc.ucla.edu/wp-content/uploads/2024/12/recall.csv")

###EXERCISE 1###
# 1. Run the random intercept model with words as the outcome and random intercepts by subj_id.


# 2. Calculate the ICC for words.


# 3. Use dplyr::mutate() to add the following to the data set:
#    a. cluster means for sleep and study (call them sleep_m and study_m)
#    b. subject-mean-centered sleep and subject-mean-centered study (call them sleep_c and study_c)


# 4. Run a multilevel model with fixed and random slopes for sleep_c,
#    interpret the coefficient, and take note of the changes in residual and 
#    random intercept variances from the random intercept model.


# 5. Add sleep_m to the model in (4), interpret its coefficient,
#    and take note of the changes in variances again.






###EXERCISE 2###
# 1. Use r2mlm() on the model in (5) in Exercise 1 
#    (fixed and random slopes of sleep_c, fixed slope for sleep_m) and interpret all output


# 2. Use r2mlm_comp() to determine whether (sleep_c + sleep_m) or (study_c + study_m) explain
#    more variation in words. Model fixed and random slopes for the subject-mean-centered predictors.