05_multivariate-graphs.Rmd

# Multivariate Graphs

**Learning objectives:**

- Incorporate multiple variables into data visualizations beyond the x and y axes via grouping them together
- Explore some of `ggplot2`'s array of customization in the context of grouping and facet-wrapping
- Use facet wrapping & facet grids to create separate graphs based on the values of different variables

## Introduction {-}
- In previous chapters, we learned to plot one to two variables on a single graph
- Here, we are looking at the `Salaries` data from the `carData` package
```{r}
library(pacman)
p_load(ggplot2, dplyr) 

data(Salaries, package="carData")

ggplot(Salaries, aes(x=yrs.since.phd,
                     y=salary)) + 
  geom_point() + 
  labs(title="Academic salary by years since degree")
```

- What if we want to visualize more than two variables?
- The `Salaries` data set has multiple columns we can plot
```{r}
Salaries %>% head()
```

## Grouping
- **Grouping** involves gathering multiple variables together and plotting them on a single graph
- You can modify different attributes of data points to reflect values of other variables 
- Ex. color, size, shape, etc (within `aes()` function for `mapping` argument in `ggplot` functions)
```{r, echo=F}
ggplot(Salaries, aes(x=yrs.since.phd,
                     y=salary,
                     color=rank)) + 
  geom_point() + 
  labs(title="Academic salary by years since degree")
```

\newpage
```{r, echo=F}
ggplot(Salaries, aes(x=yrs.since.phd,
                     y=salary,
                     color=rank, shape=sex)) + 
  geom_point() + 
  labs(title="Academic salary by years since degree")
```

- NOTE: Be mindful of how busy your graph looks.
- You can also modify the size and transparency of points.
```{r}
ggplot(Salaries, aes(x=yrs.since.phd,
                     y=salary,
                      size=yrs.service)) + 
  # alpha allows you to change transparency
  # of points
  geom_point(alpha=0.6) + 
  labs(title="Academic salary by years since degree")
```

- Here's another example of grouping with some regression lines incorporated into the graph.
- The benefits of grouping is that it also affects other `ggplot` layers you apply on top of the base `ggplot` function.
```{r}
# plot experience vs. salary with 
# fit lines (color represents sex)
ggplot(Salaries, 
       aes(x = yrs.since.phd, 
           y = salary, 
           color = sex)) +
  geom_point(alpha = .4, 
             size = 3) +
  geom_smooth(se=FALSE, 
              method = "lm", 
              formula = y~poly(x,2), 
              size = 1.5) +
  labs(x = "Years Since Ph.D.",
       title = "Academic Salary by Sex and Years Experience",
       subtitle = "9-month salary for 2008-2009",
       y = "",
       color = "Sex") +
  scale_y_continuous(label = scales::dollar) +
  scale_color_brewer(palette = "Set1") +
  theme_minimal()
```


## Faceting
- Faceting is when we have separate graphs that each correspond to different levels of a variable.
- `facet_wrap` lets you create several graphs of the same type for each level of the variable being faceted at. 
- Formula for facet wrapping is 
```{r}
ggplot(Salaries, aes(x = salary)) +
  
  geom_histogram(fill = "cornflowerblue",
                 color = "white") +
  facet_wrap(~rank, ncol = 1) +
  labs(title = "Salary histograms by rank")
```


- The `facet_grid` function is best suited for faceting against multiple variables.
- Formula for the `facet_grid` function is `row variables ~ column variables.`
```{r}
# plot salary histograms by rank and sex
ggplot(Salaries, aes(x = salary / 1000)) +
  geom_histogram(color = "white",
                 fill = "cornflowerblue") +
  facet_grid(sex ~ rank) +
  labs(title = "Salary histograms by sex and rank",
       x = "Salary ($1000)")
```

- You can also combine grouping variables together and faceting together in one graph, such as what we do with the `gapminder` data from the `gapminder` package below.
- Note that the dot in the faceting formula `.~ rank + discipline` means that there are no row variables so  `rank + discipline ~.` means no column variables.
- Also note the theme functions used in the code below are used to simplify the background color, rotate the x-axis text, and make the font size smaller
```{r}
# calculate means and standard errors by sex,
# rank and discipline

library(dplyr)
plotdata <- Salaries %>%
  group_by(sex, rank, discipline) %>%
  summarize(n = n(),
            mean = mean(salary),
            sd = sd(salary),
            se = sd / sqrt(n))

# create better labels for discipline
plotdata$discipline <- factor(plotdata$discipline,
                              labels = c("Theoretical",
                                         "Applied"))
# create plot
ggplot(plotdata, 
       aes(x = sex, 
           y = mean,
           color = sex)) +
  geom_point(size = 3) +
  geom_errorbar(aes(ymin = mean - se, 
                    ymax = mean + se),
                width = .1) +
  scale_y_continuous(breaks = seq(70000, 140000, 10000),
                     label = scales::dollar) +
  facet_grid(. ~ rank + discipline) +
  theme_bw() +
  theme(legend.position = "none",
        panel.grid.major.x = element_blank(),
        panel.grid.minor.y = element_blank()) +
  labs(x="", 
       y="", 
       title="Nine month academic salaries by gender, discipline, and rank",
       subtitle = "(Means and standard errors)") +
  scale_color_brewer(palette="Set1")
```




## Meeting Videos {-}

### Cohort 1 {-}

`r knitr::include_url("https://www.youtube.com/embed/Wz0WCFv-gOk")`

<details>
<summary> Meeting chat log </summary>

```
00:01:23	Oluwafemi Oyedele:	Hi Lydia!!!
00:01:56	Lydia Gibson:	Hey Femi!
00:05:05	Oluwafemi Oyedele:	yes, I am presenting next week!!!
00:05:58	Lydia Gibson:	Awesome!
00:15:05	Oluwafemi Oyedele:	discipline: A factor with levels A (“theoretical” departments) or B (“applied” departments)
00:52:08	Kotomi Oda:	I like this plot style
00:52:32	Lydia Gibson:	Yeah very cool
00:52:55	Tiffany Kollah:	this was very good Ken!
00:53:42	Tiffany Kollah:	thank you so much!!! 🙂
00:54:02	Oluwafemi Oyedele:	Thank you!!!
00:54:25	Kotomi Oda:	Yay!
```
</details>