richard_simualtion.R

#------------------------------------------------------------------------------
### Packages & functions

library(dplyr)
library(afex)
library(here)

# generate data from a multivariate normal distribution with known 
# correlations between values
mvrnorm <- function(n, nValues, meanValues, sdValues, corMatrix) {
  Z <- matrix(rnorm(n * nValues), nValues, n)
  t(meanValues + sdValues * t(chol(corMatrix)) %*% Z)
}
#------------------------------------------------------------------------------

sample_size <- 60
set.seed(42)

# load data
data <- read.csv("Miles_data.csv")

#-------
# NV conditions
means <- data %>% 
  filter(Condition == "NV") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., mean) %>% 
  as.numeric()

sds <- data %>% 
  filter(Condition == "NV") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., sd) %>% 
  as.numeric()

cor_mat <- data %>% 
  filter(Condition == "NV") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>% 
  cor() %>% 
  as.numeric() %>% 
  round(., 3)

cor_mat <- matrix(cor_mat, nrow = length(means), 
                 ncol = length(means), byrow = TRUE)

n <- 4
a <- means[1:4]
b <- sds[1:4]
x <- cor_nv[1:4, 1:4]
first <- mvrnorm(sample_size, 4, a, b, x)
a <- means[5:6]
b <- sds[5:6]
x <- cor_nv[5:6, 5:6]
second <- mvrnorm(sample_size, 2, a, b, x)

cond_data <- data.frame(cbind(first, second))
colnames(cond_data) <- c("Need_Sat_Pre",
                         "Need_Sat_Post", 
                         "Mood_Pre", 
                         "Mood_Post", 
                         "State_Host_Pre", 
                         "State_Host_Post")
nv_data <- cond_data %>% 
  mutate(Condition = "NV")

#-------
# NVNI conditions
means <- data %>% 
  filter(Condition == "NVNI") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., mean) %>% 
  as.numeric()

sds <- data %>% 
  filter(Condition == "NVNI") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., sd) %>% 
  as.numeric()

cor_mat <- data %>% 
  filter(Condition == "NVNI") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>% 
  cor() %>% 
  as.numeric() %>% 
  round(., 3)

cor_mat <- matrix(cor_mat, nrow = length(means), 
                  ncol = length(means), byrow = TRUE)

n <- 4
a <- means[1:4]
b <- sds[1:4]
x <- cor_nv[1:4, 1:4]
first <- mvrnorm(sample_size, 4, a, b, x)
a <- means[5:6]
b <- sds[5:6]
x <- cor_nv[5:6, 5:6]
second <- mvrnorm(sample_size, 2, a, b, x)

cond_data <- data.frame(cbind(first, second))
colnames(cond_data) <- c("Need_Sat_Pre",
                         "Need_Sat_Post", 
                         "Mood_Pre", 
                         "Mood_Post", 
                         "State_Host_Pre", 
                         "State_Host_Post")
nvni_data <- cond_data %>% 
  mutate(Condition = "NVNI")

#-------
# V conditions
means <- data %>% 
  filter(Condition == "V") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., mean) %>% 
  as.numeric()

sds <- data %>% 
  filter(Condition == "V") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>%
  sapply(., sd) %>% 
  as.numeric()

cor_mat <- data %>% 
  filter(Condition == "V") %>% 
  select(Need_Sat_Pre,
         Need_Sat_Post, 
         Mood_Pre, 
         Mood_Post, 
         State_Host_Pre, 
         State_Host_Post) %>% 
  cor() %>% 
  as.numeric() %>% 
  round(., 3)

cor_mat <- matrix(cor_mat, nrow = length(means), 
                  ncol = length(means), byrow = TRUE)

n <- 4
a <- means[1:4]
b <- sds[1:4]
x <- cor_nv[1:4, 1:4]
first <- mvrnorm(sample_size, 4, a, b, x)
a <- means[5:6]
b <- sds[5:6]
x <- cor_nv[5:6, 5:6]
second <- mvrnorm(sample_size, 2, a, b, x)

cond_data <- data.frame(cbind(first, second))
colnames(cond_data) <- c("Need_Sat_Pre",
                         "Need_Sat_Post", 
                         "Mood_Pre", 
                         "Mood_Post", 
                         "State_Host_Pre", 
                         "State_Host_Post")
v_data <- cond_data %>% 
  mutate(Condition = "V")


#------- all data
all_data <- rbind(nv_data, nvni_data, v_data)

all_data <- all_data %>% mutate(
  Age = round(rnorm(sample_size * 3, 22.5, 1.2), 0),
  Sex = sample(c("Male", "Female", "Pref not to say"), sample_size * 3, replace = TRUE, 
               prob = c(.475, .475, .05)),
  Videogame_experience = sample(data$Vidgame_experience, sample_size * 3, 
                                replace = TRUE),
  Weekly_gaming = sample(data$Weekly_gaming, sample_size * 3, replace = TRUE)
)

write.csv(all_data, "simulated_data.csv", row.names = FALSE)