Plotting in R: Part III

class: center, middle, inverse, title-slide

.title[
# Plotting in <strong>R</strong>: Part III
]
.subtitle[
## <a href="https://eligurarie.github.io/EFB654/">EFB 654: R and Reproducible Research</a>
]
.author[
### Elie Gurarie
]
.date[
### <strong>February 23, 2026</strong>
]

---

class: inverse

## Today

- **Layouts** — `mfrow` and `layout()`
- **Exporting figures** — `png()`, `pdf()`, `jpeg()`, and knitr chunk options
- **Brief intro to `ggplot2`**

Using the WDI data from [last time](PlottingPartII_lab.html)

---

## Setup — load data & colors

``` r
wdi <- read.csv("WDI.csv")
wdi$region <- factor(wdi$region)
cols <- c("East Asia & Pacific"        = "steelblue",
          "Europe & Central Asia"      = "forestgreen",
          "Latin America & Caribbean"  = "goldenrod",
          "Middle East & North Africa" = "darkorange",
          "North America"              = "mediumpurple",
          "South Asia"                 = "firebrick",
          "Sub-Saharan Africa"         = "sienna")
```

---
class: inverse

# 1. Controlling layouts

---

## `par(mfrow = ...)`

.pull-left[

``` r
par(mfrow = c(2, 1))
boxplot(GDP/1000 ~ region, data = wdi,
        log = "y", las = 1, col = cols)
boxplot(LifeExp ~ region, data = wdi,
        las = 1, col = cols)
```

- `mfrow = c(nrow, ncol)` — fill by **row**
- `mfcol = c(nrow, ncol)` — fill by **column**
]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/mfrow-basic-1.png" alt="" style="display: block; margin: auto;" />
]

---

## Prettier: `mar` and `oma`

.pull-left[.small[

``` r
par(mfrow = c(2,1), 
    mar = c(0,4,1,2),   # margins per panel
    oma = c(3,0,2,0),   # outer margins
    cex.axis = 0.8, 
    tck = 0.01, bty = "l", mgp = c(2,.25,0))

boxplot(GDP/1000 ~ region, data = wdi, 
        log = "y", las = 1, col = cols, 
        xlab = "", xaxt = "n")
boxplot(LifeExp ~ region, data = wdi, 
        las = 1, col = cols)
title("Global Indicators across regions",
      outer = TRUE)
```
]]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/mfrow-pretty-1.png" alt="" style="display: block; margin: auto;" />
]

---

## Margin anatomy

.pull-left[

| parameter | what it controls |
|-----------|-----------------|
| `mar` | margins around **each** panel (bottom, left, top, right) |
| `oma` | **outer** margins around entire figure |
| `xaxt = "n"` | suppress x-axis (draw your own later) |

`title(..., outer = TRUE)` — title in outer margin
]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/margin-diagram-1.png" alt="" style="display: block; margin: auto;" />
]

---

## Fixing long labels

.pull-left[.small[

``` r
par(mfrow = c(2,1), 
    mar = c(0,4,1,2), oma = c(3,0,2,0),
    cex.axis = 0.8, tck = 0.01, bty = "l",
    mgp = c(2,.25,0))

boxplot(GDP/1000 ~ region, data = wdi, 
        log = "y", las = 1, col = cols, 
        varwidth = TRUE, 
        xlab = "", xaxt = "n")
boxplot(LifeExp ~ region, data = wdi, las = 1, 
        varwidth = TRUE, col = cols, xaxt = "n")
title("Global Indicators across regions",
      outer = TRUE)

regions <- levels(factor(wdi$region))
region_labels <- gsub("&", "&\n", regions,
                       fixed = TRUE)
mtext(region_labels, side = 1, 
      at = 1:7, line = 0, padj = 1)
```
]]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/mfrow-labels-1.png" alt="" style="display: block; margin: auto;" />
]

---

## Key tricks used

- `varwidth = TRUE` — box width proportional to sample size
- `xaxt = "n"` — suppress automatic x-axis labels
- `gsub("&", "&\n", ...)` — insert newlines to break long labels
- `mtext()` — manual axis label placement

---
class: inverse

# `layout()`

---

## `layout()` — flexible panel arrangements

.pull-left[

``` r
layout(rbind(c(1, 2), c(1, 3)))
layout.show(3)
```

<img src="PlottingPartIII_slides_files/figure-html/layout-demo-1.png" alt="" style="display: block; margin: auto;" />
]

.pull-right[
- `layout()` takes a **matrix** of panel numbers
- Panel 1 spans both rows (left column)
- Panels 2 & 3 stack on the right
- `layout.show(n)` — preview the arrangement
]

---

## Complex layout in action

[See complete code in lab]

---
class: inverse

# Quick aside: wrapping code in functions

---

## Bundle plotting code into functions

.pull-left[.small[

``` r
setPars <- function(){
  par(cex.lab = 1.2, las = 1,
    mgp = c(2, .25, 0), tck = 0.01, 
    bty = "l", mar = c(0,4,0,2), 
    oma = c(4,0,4,0), xpd = NA)
}

plotWDI <- function(){
  cont_col <- cols[match(wdi$region,
                         levels(wdi$region))]
  cex_pop <- sqrt(wdi$Population / 
      max(wdi$Population, na.rm = TRUE)) * 15
  plot(wdi$GDP/1e3, wdi$LifeExp, log = "x",
       ylab = "Life expectancy (years)",
       xlab = "GDP per capita (1000 USD)",
       type = "n")
  points(wdi$GDP/1e3, wdi$LifeExp,
         cex = cex_pop + .5, pch = 21,
         col = cont_col,
         bg = alpha(cont_col, .7))
  legend("bottomright", title = "Region",
         legend = levels(wdi$region),
         pt.bg = cols, col = "grey40",
         pch = 21, pt.cex = 1.8, bty = "n")
}

boxplotWDI <- function(){
  boxplot(GDP/1000 ~ region, data = wdi, 
          log = "y", las = 1, varwidth = TRUE, 
          col = alpha(cols, .7), border = cols, 
          xlab = "", xaxt = "n")
  boxplot(LifeExp ~ region, data = wdi, las = 1, 
          varwidth = TRUE, col = alpha(cols, .7),
          border = cols, xaxt = "n",
          xlab = "Region")
}
```
]]

.pull-right[

- No arguments, no return values
- Just **do the thing**
- Functions only plot — setup is separate

Then the complex layout becomes:

``` r
layout(rbind(c(1,2), c(1,3)))
setPars()
plotWDI()
boxplotWDI()
```

Three lines instead of 30+
]

---

``` r
layout(rbind(c(1,2), c(1,3)))
setPars(); plotWDI(); boxplotWDI()
```

---
class: inverse

# 2. Exporting figures to files

---

## The pattern: open → plot → close

``` r
device_function("filename.ext", width = ..., height = ...)
  # ... plotting code ...
dev.off()                          # close and write file
```

- Nothing appears on screen — output goes straight to file
- **Forgetting `dev.off()`** = most common mistake
- The file stays open and incomplete

---

## PDF vs. PNG

| | PDF | PNG |
|---|---|---|
| Type | **Vector** | **Raster** (bitmap) |
| Stores as | Math descriptions | Grid of pixels |
| Scaling | Perfect at any size | Degrades if enlarged |
| Best for | Publications, LaTeX | Web, Word, presentations |
| Resolution | N/A | Critical — use `res =` |

---

## Exporting to PDF

``` r
pdf("WDImegaplot.pdf", width = 12, height = 6)   # inches

layout(rbind(c(1,2), c(1,3)))
  setPars()
  plotWDI()
  boxplotWDI()

dev.off()
```

- `width` / `height` always in **inches** (default 7 × 7)
- Plotting in a loop → **multi-page PDF**

---

## Exporting to PNG

``` r
png("WDImegaplot.png", width = 2400, height = 1200, res = 200)   # pixels

layout(rbind(c(1,2), c(1,3)))
  setPars()
  plotWDI()
  boxplotWDI()

dev.off()
```

- `width` / `height` in **pixels** (by default)
- `res` = dots per inch — controls text/point size relative to canvas
- Effective size = `width / res` → `2400/200 = 12"` wide
- Print quality: `res = 300`; screen: `res = 96`–`150`

---

## Controlling size in R Markdown

````
```{r myplot, fig.width = 8, fig.height = 5, dpi = 200}`
# ... plotting code ...
```
````

- `fig.width` / `fig.height` — **inches** (like PDF)
- `dpi` — resolution of rendered PNG (knitr default: 72)
- Set defaults globally in setup chunk:

``` r
knitr::opts_chunk$set(fig.width = 7, fig.height = 5, dpi = 150)
```

---

## `jpeg()` — when to use

``` r
jpeg("myplot.jpg", width = 800, height = 600, 
     res = 150, quality = 90)
  # ... plotting code ...
dev.off()
```

- `quality` 0–100 (lower = smaller file, more artifacts)
- **Avoid** for plots with sharp lines & text — artifacts visible
- OK for photographic content or when file size matters

---

## Device summary

| Function | Format | Size units | Key arg |
|----------|--------|-----------|---------|
| `pdf()` | Vector PDF | inches | `width`, `height` |
| `png()` | Raster PNG | pixels | `res` (dpi) |
| `jpeg()` | Raster JPEG | pixels | `quality` |
| knitr chunk | (PNG) | inches | `fig.width`, `dpi` |

---
class: inverse

# 3. Introduction to `ggplot2`

---

## What is `ggplot2`?

- Created by Hadley Wickham (2005)
- Based on *The Grammar of Graphics* (Wilkinson 1999)
- Every graphic = **data** + **aesthetics** + **geometries**
- You *declare* what you want; `ggplot2` figures out the rendering
- Result is a single object you can inspect, modify, print

---

## The key constraint

**Every variable must be a column in a data frame**

- Can't pass loose vectors like in `plot(x, y)`
- Sometimes inconvenient, but enforces good data discipline

``` r
library(ggplot2)
```

---

## Basic scatter plot

.pull-left[

``` r
ggplot(data = wdi, 
       aes(x = GDP/1e3, y = LifeExp)) +
  geom_point()
```

`aes()` maps **variables** to **visual channels**

| aesthetic | visual property |
|-----------|----------------|
| `x`, `y`  | position |
| `color`   | outline color |
| `fill`    | fill color |
| `size`    | point size |
| `alpha`   | transparency |
| `shape`   | point character |
]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/gg-basic-1.png" alt="" style="display: block; margin: auto;" />
]

---

## Common `geom_*` functions

| geom | draws |
|------|-------|
| `geom_point()` | scatter plot |
| `geom_line()` | lines |
| `geom_smooth()` | trend + confidence band |
| `geom_boxplot()` | box-and-whisker |
| `geom_histogram()` | histogram |
| `geom_bar()` / `geom_col()` | bar chart |
| `geom_text()` | text labels |

---

## Adding dimensions is quick

.pull-left[

``` r
ggplot(wdi, aes(x = GDP/1e3, y = LifeExp, 
                color = region, 
                size  = Population/1e6)) +
  geom_point(alpha = 0.7) +
  scale_x_log10() +
  theme_bw()
```

- 4 variables in ~4 lines
- Legends generated **automatically**
- Compare to the manual `match()` + `legend()` work in base R
]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/gg-multi-1.png" alt="" style="display: block; margin: auto;" />
]

---

## BUT — customization is hard

.pull-left[
Moving a legend, adjusting ticks, mixing fonts — all require `theme()` with ~100 arguments or extension packages.

``` r
theme(
  legend.position   = c(0.15, 0.85),
  legend.background = element_rect(
      fill = "white", color = "grey70"),
  axis.text         = element_text(size = 11),
  panel.grid.minor  = element_blank(),
  plot.title.position = "plot"
)
```
]

.pull-right[
In base R the same thing is a couple `par()` arguments.

> Customizing a ggplot is one of the better uses of Large-Language Models
]

---

## Faceting — extremely powerful

.pull-left[.small[

``` r
ggplot(wdi, aes(x = GDP/1e3, y = LifeExp, 
                size = Population/1e6)) +
  geom_point(alpha = 0.6, color = "steelblue") +
  geom_smooth(method = "lm", color = "firebrick",
              linewidth = 0.7) +
  scale_x_log10() +
  facet_wrap(~ region)
```
]]

.pull-right[
<img src="PlottingPartIII_slides_files/figure-html/gg-facet-1.png" alt="" style="display: block; margin: auto;" />
]

One line: `facet_wrap(~ region)` — automatic panels, shared scales, per-panel trend lines.

---

## `ggplot2` vs. base R

| | base R | ggplot2 |
|---|---|---|
| Learning curve | Steeper for basics | Steeper for customization |
| Multi-variable | Manual (but explicit) | Fast (auto legends/scales) |
| Faceting | Manual `layout()` | `facet_wrap()` / `facet_grid()` |
| Fine control | Excellent | Difficult |
| Default aesthetics | Minimal | Ugly (IMO) |
| Data requirement | Anything | Data frame required |

The two are **complementary**: `ggplot2` for exploration & panels; base R for precise, publication-ready control.