functions_exercises_answers.qmd

---
title: "Custom Functions Exercises"

date: "today"

format: html
editor: visual
---

## A. Workshop exercises

### A.1. Load tidyverse

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

### A.2. Set themes

Run this chunk of code to set custom themes (`theme_bar` and `theme_line`) for our later graphics.

```{r}
theme_bar <- theme_minimal() +
    theme(strip.text = element_text(size = 10.5, face = "bold", 
                                  color = "grey30", margin = margin(b=10)),
        axis.text = element_text(size = 10),
        axis.title.y = element_blank(),
        axis.title.x = element_text(margin = margin(t=10), color = "grey30"),
        axis.line.x = element_line(color = "grey30"),
        panel.grid = element_blank(),
        panel.grid.major.x = element_line(color = "grey80", linetype = "dotted"),
        panel.spacing = unit(1,"cm"),
        legend.title = element_blank())

theme_line <- theme_minimal() +
  theme(axis.text = element_text(size = 10),
        axis.title.x = element_blank(),
        axis.title.y = element_text(margin = margin(r=10), color = "grey30"),
        panel.grid = element_blank(),
        panel.grid.major.y = element_line(color = "grey80"),
        legend.title = element_blank())
```

### A.3. Load data

Today we'll be using a fake data set that includes student athlete GPAs from 2014 to 2024, broken down by sport and student status, sex, and race/ethnicity.

*Read in the "college_student_athletes.csv" file*

```{r}
athletes <- read_csv("college_student_athletes.csv")

athletes
```

### A.4. Aggregate

For future analysis we want four new data frames that include a **count of athletes** and their **average GPAs**:

1.  by year and sport

2.  by year, sport, and status

3.  by year, sport, and sex

4.  by year, sport, and race/ethnicity

*HINT: Use* `group_by()` *and* `summarize()`

```{r}
# by year and sport
year_sport <- athletes |> 
  group_by(Year, Sport) |> 
  summarize(n_athletes = n(),
            avg_GPA = mean(GPA))
year_sport

# by year, sport, and status/sex/race
agg <- function(myvar) {
  athletes |> 
    group_by(Year, Sport, {{ myvar }}) |> 
    summarize(n_athletes = n(),
              avg_GPA = mean(GPA))
}

# run function
by_status <- agg(Status)
by_sex <- agg(Sex)
by_race <- agg(Race)

```

### A.5. Bar charts

Using a function, build a set of bar charts that show the number of athletes who played each sport in 2024 faceted by status, sex, then race/ethnicity. Your function should also allow you to highlight one or more sports using fill color.

```{r}

# build function
fill_bars <- function(mydata, mysports, myvar){
  mydata |> 
  filter(Year == 2024) |> 
  ggplot(aes(n_athletes, 
             reorder(Sport, n_athletes),
             fill = Sport %in% mysports)) +
  geom_col(show.legend = F) + 
  facet_wrap(vars({{ myvar }})) +
  scale_fill_manual(values = c("lightgrey","steelblue")) +
  theme_bar
}

# run function
fill_bars(by_status, c("Lacrosse","Tennis"), Status)
fill_bars(by_sex, c("Baseball","Soccer"), Sex)
fill_bars(by_race, "Swimming", Race)

```

### A.6. Line chart

Create a line chart using the `year_sport` data frame that shows how average GPAs have changed over time by sport. Then grey out all the lines to prepare to highlight in the next step.

```{r}
# create color and reference vectors 
mycolors <- rep("lightgrey", 7) 

mysports <- year_sport |> 
  pull(Sport) |> 
  unique()

# plot 
year_sport |>   
  ggplot(aes(Year, avg_GPA, color = Sport)) +   
  geom_line(linewidth = 1) + 
  scale_y_continuous(expand = c(0,0), limits = c(2.7,3.5)) +
  scale_x_continuous(breaks = seq(2014,2024,2)) +
  scale_color_manual(values = mycolors) +
  theme_line
```

Build a function that highlights one line at a time, keeping all other lines grey.

```{r}

# build function 
color_lines <- function(x) {

  mycolors[x] <- "coral2"
  
  myplot <- year_sport |>     
    ggplot(aes(Year, avg_GPA, color = Sport)) +     
    geom_line(linewidth = 1) +     
    geom_line(data = year_sport |> filter(Sport == mysports[x]),
              linewidth = 1.5) +     
  scale_y_continuous(expand = c(0,0), limits = c(2.7,3.5)) +
  scale_x_continuous(breaks = seq(2014,2024,2)) +
  scale_color_manual(values = mycolors) +
  theme_line
  
  ggsave(plot = myplot,
         filename = paste0("sport_line_",x,".png"),
         width = 6, height = 4,
         bg = "white")
  
  myplot
}  

# run function 
map(1:7, color_lines)
```

## B. On your own

Use the aggregated data frames you created in section A.4. to build plots showing the breakdown of average GPA:

-   faceted by status in 2024

-   faceted by sex in 2020

-   faceted by race in 2015

*Create the three plots listed above (they can be horizontal bar charts, but you might also try a dot plot using* `geom_point()` *instead). Test your code with the first one, then build a function to create the other two. HINT: Your function inputs should be the aggregated data frame, year, and grouping variable* (`Status` / `Sex` / `Race`).

```{r}

# test it
by_status |> 
  filter(Year == 2024) |> 
  ggplot(aes(avg_GPA, reorder(Sport, avg_GPA))) +
  geom_point() +
  facet_wrap(~Status) + 
  ggtitle(2024)

# build function
plot_points <- function(mydf, myyear, myvar) {
  mydf |> 
    filter(Year == myyear) |> 
    ggplot(aes(avg_GPA, reorder(Sport, avg_GPA))) +
    geom_point() +
    facet_wrap(vars({{ myvar }})) +
    ggtitle(myyear)
}

# run function
plot_points(by_sex, 2020, Sex)
plot_points(by_race, 2015, Race)
```