---
title: "Lab-1"
output: html_document
---
# Instructions
- You will do your work in this `Quarto` document but will submit all your answers on Canvas
- *very important*: As you move through the document, make sure to run all code chunks (grey bits) that already have code in them
- Write your own code in the empty code chunks
- Useful shortcuts:
- to run all the code in a specific code chunk, press the green right-facing triangle at the top right of the code chunk
- to run all *prior* code chunks, press the downward-facing gray triangle at the top right of the code chunk
```{r load-libraries, warning=FALSE, message=FALSE}
library(tidyverse) # Load ggplot2, dplyr, and all the other tidyverse packages
library(babynames) # loads babynames data, needs to install if not yet installed
library(socviz) # loads socviz data, needs to install if not yet installed
library(juanr) # install if not yet installed
```
# Task 1: Baby Names
The `babynames` package includes data from the Social Security Administration on how often a name was given to a baby between 1880-2015.
Run the code chunk below to create `baby_subset`, a dataset of babies named "Angel".
```{r}
baby_subset <- babynames %>%
filter(name == "Angel")
# look at the data
head(baby_subset)
```
Using `baby_subset`, make a *time-series* that shows how often the name we picked was used for babies categorized as male and as female over time. (Hint: map the `sex` variable to the color aesthetic; also, the `n` variable shows the count per year).
Make the plot below:
```{r}
```
**Question 1**: Roughly speaking, what happens to the use of the name from the 1990s onwards?
Now, make a similar graph for your name or a name of your choosing. Alter the code chunk below so that `baby_subset` contains babies with your chosen name instead of "Angel". If `baby_subset` looks empty, make sure the name is spelled correctly (e.g., "Juan" instead of "juan"). If you're still struggling, try a different name!
```{r mckinley-subset}
baby_subset <- babynames %>%
filter(name == "Angel")
# look at the data
head(baby_subset)
```
Now make the same plot from earlier below:
```{r}
```
**Question 2: Briefly describe the trend you observe with your chosen name over time.**
Finally, let's look at the most popular names in 1945. Run the code below:
```{r}
top_names = babynames %>%
filter(year == 1945) %>%
group_by(sex) %>%
top_n(n = 10, wt = prop) %>%
select(sex, name, prop) %>%
mutate(name = fct_reorder(name, prop))
```
Using `top_names`, make a barplot that puts the baby's name on the y-axis, and the proportion of babies with that name (the variable is called `prop`) on the x-axis. Map the sex variable to the fill aesthetic to create separate bars for male and female babies.
```{r}
```
**Question 3: Roughly speaking, where is the biggest drop off in name popularity among female names?**
# Task 2: Childcare costs
Run the code chunk below to generate `childcare_four`, a dataset of childcare costs for every county in four US states.
```{r}
childcare_four = childcare |>
filter(state %in% c("Mississippi", "Virginia", "California", "Hawaii")) |>
select(state, county, infant_price)
childcare_four
```
Using `childcare_four`, make a boxplot plot of the distribution of infant childcare prices across states (hint: the state's name should go on the y-axis).
```{r}
```
**Question 4: Generally speaking, which state has the most outliers in terms of childcare costs?**
# Task 3: The Movies
Next, let's look at some data about movies. These data are about a selection of films and info about how much they cost, how much they made in theaters, and what rating they received from reviewers. Run the code below:
```{r clean-movie-data, message=FALSE, warning=FALSE}
# read data
?movies # look at the codebook
movies =
movies %>%
# exclude films made before the 60s
filter(year >= 1960)
head(movies)
```
Using `movies`, make a scatterplot that presents a movie's budget (x-axis) against how much it grossed in theaters (y-axis). Color the points by which `decade` the movie was made in. Add titles to the x and y axis and a theme.
Make the plot below:
```{r budget}
```
**Question 5: right click the graph --\> "save image as" --\> and save it somewhere on your computer. You will need to submit the plot!**
Look at that outlier point in top-left part of the plot (should have the color for "1970s"). What movie is this?
Make a new plot that adds labels to each of the points so that you can figure out what that movie is. Hint: look at how we did this with the health and wealth plot from class with `gapminder`.
Make the plot below (hint: if hard to see, press the icon with an arrow and a box at the top right of the image):
```{r}
```
**Question 6: What's the outlier movie?**
**Question 7: What, if anything, does the movie's position on the graph tell us about how successful it was?**
Movies vary so much in length (i.e., their duration in minutes). Around how long are most movies? Make a histogram of `duration` below:
```{r ratings-year}
```
**Question 8: Roughly, about how long are most movies (where is the peak of the distribution)?**
Finally, let's look at who are the most/least *profitable* thriller movie directors. First, we'll make a new dataset that only looks at thriller movies, calculates each movie's profit (`gross` - `budget`) and finds the average profit for each director. To make this more manageable, let's look at directors who made at least 2 movies.
Run the code below to create `directors_profit`, a dataset of thriller movie directors:
```{r summarize-profits-director}
directors_profit = movies %>%
# only look at horror movies
filter(genre1 == "Thriller" | genre2 == "Thriller" | genre3 == "Thriller") %>%
# calculate profit
mutate(profit = gross - budget) %>%
# group by director name
group_by(director) %>%
# calculate number of movies each director made, average profit, average rating
summarise(n_movies = n(),
profit = mean(profit)) %>%
# include directors who made more than 2 movies
filter(n_movies > 2)
```
Using `directors_profit` make a barplot that has the `director` on the y-axis and `profit` on the x-axis.
Make the plot below:
```{r plot-director-profits}
```
- **Question 9: Who is the least profitable director?**
# Task 4: Pokemon
First we'll look at data on pokemon. Let's load it below and look at it.
```{r load-pokemon}
pokemon
?pokemon
```
Using `filter()`, tell me which Pokemon:
- is taller than a meter
- weighs more than 200 times its height
- has a type 1 of "ground"
- is a generation 3 pokemon
```{r}
```
**Q10: Who's that Pokemon?**
Now, look at Pokemon that:
- have a higher defense than attack
- have a capture rate of 3
- are legendary
```{r}
```
**Q11: How many pokemon fit this description?**
Next, look at Pokemon that are:
- grass for type 1 and psychic for type 2, or grass for type1 and flying for type2
- better at speed than defense
- have an HP level above 50
```{r}
```