import libraries

library(tidyverse)
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.3.6      ✔ purrr   0.3.4 
## ✔ tibble  3.1.8      ✔ dplyr   1.0.10
## ✔ tidyr   1.2.1      ✔ stringr 1.4.1 
## ✔ readr   2.1.2      ✔ forcats 0.5.2 
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
library(moderndive)
library(skimr)
library(gapminder)
evals_ch5 <- evals %>%
  select(ID, score, bty_avg, age)

LC5.1

evals_ch5 %>% select(age, score) %>% skim()
Data summary
Name Piped data
Number of rows 463
Number of columns 2
_______________________
Column type frequency:
numeric 2
________________________
Group variables None

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
age 0 1 48.37 9.80 29.0 42.0 48.0 57.0 73 ▅▆▇▆▁
score 0 1 4.17 0.54 2.3 3.8 4.3 4.6 5 ▁▁▅▇▇ 
evals_ch5 %>% 
  get_correlation(formula = score ~ age)
ggplot(evals_ch5, aes(x = age, y = score)) +
  geom_jitter() +
  geom_smooth(method = "lm", se = FALSE)
## `geom_smooth()` using formula 'y ~ x'

Score is negatively correlated with age.

LC5.2

get_regression_table(lm(score ~ age, data = evals_ch5))

this slope and intercept are the same as the line from geom_smooth()

LC5.3

get_regression_points(lm(score ~ age, data = evals_ch5))
get_regression_points(lm(score ~ age, data = evals_ch5)) %>% summarise(ssr = sum(residual**2), mse=mean(residual**2))

LC5.4

gapminder %>% filter(year==2007) %>% 
  ggplot() +
    geom_histogram(aes(gdpPercap)) +
    facet_wrap(~continent)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

gapminder %>% filter(year==2007) %>% 
  ggplot() +
    geom_boxplot(aes(continent, gdpPercap))

gapminder %>%
  filter(year==2007) %>%
  group_by(continent) %>%
  summarize(median = median(gdpPercap), 
            mean = mean(gdpPercap))

Africa has by far the lowest gdpPercap. Asia has the most spread.

LC5.5

get_regression_table(lm(gdpPercap ~ continent, data = gapminder %>% filter(year==2007)))

The baseline y intercept for Africa is equal to the mean we found earlier. Actually, all of the estimates + the africa estimate equal their mean from earlier. Not that useful though.

LC5.6

get_regression_points(lm(lifeExp ~ continent, data = gapminder %>% filter(year==2007)), ID = "country") %>% arrange(residual) %>% head()

These countries life expectancy is farthest below the average for their continent

LC5.7

get_regression_points(lm(lifeExp ~ continent, data = gapminder %>% filter(year==2007)), ID = "country") %>% arrange(desc(residual)) %>% head()

These countries life expectancy is farthest above the average for their continent

LC5.8

Red: -0.5, -1.5, 0.5 -> SSR = 2.75 Green: 0, -1.5, 2 -> SSR = 6.25 Blue: 0.5, -1, 0.5 -> SSR = 1.5