Lab 1 Question 2.qmd

---
title: "Untitled"
format: html
editor: visual
---

## Question 2

```{r}
library(tidyverse)
library(purrr)
library(readr)
```

```{r}
oscars_data <- read_csv("Oscars-demographics-DFE.csv")
```

```{r Warmups}
#1 
oscars_data |>
  select(movie, award, year_of_award) |>
  count(movie) |>
  slice_max(order_by = n,n = 1)


#2
oscars_data |>
  filter(award == "Best Actress") |>
  select(year_of_award, person) |>
  mutate(person = str_replace(person, "(?s) .*", "")) |>
  count(person) |>
  slice_max(order_by = n, n = 1)

#3 - Brandon's Code
oscars_data |> 
  mutate(birthplace = ifelse(birthplace=="New York City", "Ny", birthplace),
         region = sub(".*, ", "", birthplace)) |>
  count(region) |>
  filter(n == max(n))
```

```{r Bootstrapping - Tidy}
#What is an approximate 95% confidence interval for percent of "Big 5 Award" award winners who are not white?

prop_function <- function(race_data) {
  
white_winners <- race_data |>
  filter(race_ethnicity == "White") |>
  count() |>
  as.numeric()

nonwhite_winners <- race_data |>
  filter(race_ethnicity != "White") |>
  count() |>
  as.numeric()

return(nonwhite_winners / white_winners)

}

#old_sample <- oscars_data |>
#  select(race_ethnicity)

#new_sample <- sample(old_sample, replace = TRUE)

#oscars_data |>
#  select(race_ethnicity) |>
#  map_int(.f = prop_function)

```


```{r Copied Code from Stack Overflow for Testing}


race_ethnicity <- oscars_data |>
  select(race_ethnicity)

# number of bootstrap replicates
B <- 100

# create empty storage container
result_vec <- vector(length=B)

for(b in 1:B) {
    # draw a bootstrap sample
    this_sample <- sample(race_ethnicity, size=length(race_ethnicity), replace=TRUE)

    # calculate your statistic
    m <- prop_function(this_sample)

    # save your calucated statistic
    result_vec[b] <- m
}

# then probably draw a histogram of your bootstrapped replicates
hist(result_vec)

# get 95% confidence interval
result_vec <- result_vec[order(result_vec)]
lower_bound <- result_vec[round(0.025*B)]
upper_bound <- result_vec[round(0.0975*B)]

```