# This chapter uses packages that takes a few minutes to download on Google Colab. 
# Run this first so it's ready by the time you need it
install.packages("ggformula")
install.packages("dplyr")
library(ggformula)
library(dplyr)

studentdata <- read.csv("https://raw.githubusercontent.com/smburns47/Psyc158/main/studentdata.csv")

thumb_model <- #use lm() here with the variable "Thumb" in dataset "studentdata"
studentdata$Thumb_residuals <- #use resid() or thumb_model$residuals here 
head(studentdata)

gf_histogram( ~ Thumb_residuals, data = studentdata) %>% 
  gf_vline(., xintercept = mean(studentdata$Thumb_residuals))

not_the_mean <- 50

studentdata$Thumb_bad_residuals <-  studentdata$Thumb - not_the_mean
gf_histogram( ~ Thumb_bad_residuals, data = studentdata) %>% 
  gf_vline(., xintercept = mean(studentdata$Thumb_bad_residuals), color = "red")

gf_histogram( ~ Thumb_residuals, data = studentdata) %>% 
  gf_vline(., xintercept = mean(studentdata$Thumb_residuals))

sum(studentdata$Thumb_residuals)

sum(studentdata$Thumb_residuals^2)

install.packages("supernova")
library(supernova)

supernova(thumb_model)

var(studentdata$Thumb_residuals)

#same as dividing the sum of squares by N-1
sum(studentdata$Thumb_residuals^2) / (length(studentdata$Thumb) - 1)

supernova(#YOUR CODE HERE)

12970.4846566879/156

#same as the variance in the residuals
var(studentdata$Thumb_residuals)

sd(studentdata$Thumb_residuals)

#same as taking the square root of variance
sqrt(var(studentdata$Thumb_residuals))

#same as dividing the sum of squares by N-1 and then taking the square root
sqrt(sum(studentdata$Thumb_residuals^2) / (length(studentdata$Thumb) - 1))

gf_histogram( ~ Thumb_residuals, data = studentdata) %>% 
  gf_vline(., xintercept = mean(studentdata$Thumb_residuals))

# find the standard deviation of studentdata$Thumb

thumb_sd <- 

gf_histogram( ~ Thumb, data = studentdata) %>% 
  gf_vline(., xintercept = mean(studentdata$Thumb), color = 'blue') %>%
  gf_vline(., xintercept = 64, color = 'red')

studentdata$Thumb_rescaled <- (studentdata$Thumb - mean(studentdata$Thumb)) / sd(studentdata$Thumb)
studentdata$Thumb[11] # value of Zelda's thumb length, in original units
studentdata$Thumb_rescaled[11] # new value of Zelda's thumb length, in new units

studentdata$Thumb[3] 
studentdata$Thumb_rescaled[3]