library(broom)
library(tidyverse)
library(janitor)
healthdata <- read.spss("Health_LISS_Core_Study_Wave_12_2020_data_plus_background.sav", to.data.frame = TRUE)

# Create healthdata_s
healthdata_s <- healthdata %>%
    select(ch19l016, ch19l017, ch19l004, ch19l001, ch19l002, nettoink)
summary(healthdata_s)
healthdata_s %>% View()

# add a new category called BMI
healthdata_s <- healthdata_s %>%
  mutate(BMI = (ch19l017 / (ch19l016 / 100)^2))

# create histogram
healthdata_s %>%
  ggplot() +
  geom_histogram(aes(x = BMI))

# only show people with a BMI of 14 or higher and 50 or lower
healthdata_s <- healthdata_s %>%
  filter(BMI >= 14 & BMI <= 50)

# show mean and sd and total amount
healthdata_s %>%
summarise(mean = mean(BMI), sd = sd(BMI))

mean(healthdata_s$BMI, na.rm = TRUE)
sd(healthdata_s$BMI, na.rm = TRUE)
nrow(filter(healthdata_s, BMI != "NA"))

# mutate health variable to numbers
healthdata_s <- healthdata_s %>%
  mutate(generalhealth = case_when(
    ch19l004 == "poor" ~ 1,
    ch19l004 == "moderate" ~ 2,
    ch19l004 == "good" ~ 3,
    ch19l004 == "very good" ~ 4,
    ch19l004 == "excellent" ~ 5))

# histogram variable health
healthdata_s %>%
  ggplot() +
  geom_histogram(aes(x = generalhealth))

# scatterplot
healthdata_s %>%
  ggplot(aes(x = BMI, y = generalhealth)) +
  geom_point() +
  geom_smooth(method = "lm", se= FALSE)

# OLS model
model <- healthdata_s %>%
  lm(generalhealth ~ BMI, data = .)
model
summary(model)

# standardizing part
model_2 <- healthdata_s %>%
  lm(scale(generalhealth) ~ scale(BMI), .)

summary(model_2)

# standardized values scatterplot
healthdata_s %>%
  ggplot(aes(x = BMI, y = generalhealth)) +
  geom_point() +
  geom_smooth(method = "lm", se= FALSE)