the ggplot glow-up

class: center, middle, inverse, title-slide

# the ggplot glow-up
## making lovely data visualizations in R
### Ludmila Janda
### Amplify
### 2019/09/19

---

# Amplify

![](images/amplify_science.jpg)

---

[Glow-up definition](https://twitter.com/zhiiiyang/status/1135743078881304576)

![](images/hgu.png)

---

### Motivation

![](images/tufte_tweet.png)

---

### Motivation

![](images/tufte1.png)

---

### Motivation

![](images/blocked_by_tufte2.png)

---

### Motivation

- One of the great powers of R is visualization

- Yet graphs like this persist

![](ny_hackr_talk_2019_09_19_files/figure-html/bad_graph-1.png)

---

```r
ggplot(diamonds, aes(factor(cut), price)) + 
  geom_boxplot() +
  labs(title = "Diamond Prices By Cut") +
  theme_minimal()
```

![](ny_hackr_talk_2019_09_19_files/figure-html/theme-1.png)

---

### Make your own theme!

```r
theme_diamonds <- function(x) {
  theme(text = element_text(family = "Luminari"),
        axis.text = element_text(size = 10, color = "skyblue"),
        axis.title.x = 
          element_text(size = 12, color = "skyblue", vjust = -3),
        axis.title.y = 
          element_text(size = 12, color = "skyblue", vjust = 6),
        axis.line.x = element_line(color  = "gold"), 
        axis.line.y = element_line(color  = "gold"), 
        axis.ticks = element_line(color  = "gold"),
        plot.title = 
          element_text(size = 14, face = "bold", color = "skyblue"),
        plot.caption = element_text(color = "skyblue"),
        plot.margin = margin(t = 0.5, r = 0.5, b = 0.5, l = 0.5, "cm"),
        panel.background = element_rect(fill = "white"),
        panel.grid.minor = element_blank())
}
```

---

![](ny_hackr_talk_2019_09_19_files/figure-html/use_theme_diamonds-1.png)

---

```r
library(scales)
ggplot(diamonds, aes(cut, price)) + 
  geom_boxplot() +
  scale_y_continuous(label = comma) +
  labs(x = "Cut", y = "Price", title = "Diamond Prices By Cut") +
  theme_diamonds()
```

![](ny_hackr_talk_2019_09_19_files/figure-html/scales-1.png)

---

```r
ggplot(diamonds, aes(cut, price)) + 
  geom_dotplot(binaxis = "y", stackdir = "center", stackratio = 0.5, 
               dotsize = 0.5, binwidth = 30, alpha = 0.5) +
  scale_y_continuous(label = comma) +
  labs(x = "Cut", y = "Price", title = "Diamond Prices By Cut") +
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/dotplot-1.png)

---

```r
ggplot(diamonds, aes(cut, price, color = cut)) + 
  geom_dotplot(binaxis = "y", stackdir = "center", stackratio = 0.5, 
               dotsize = 0.5, binwidth = 30, alpha = 0.5) +
  gghighlight::gghighlight(cut == "Ideal") + 
  scale_y_continuous(label = comma) +
  labs(x = "Cut", y = "Price", title = "Diamond Prices By Cut") +
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/dotploth-1.png)

---

```r
ggplot(diamonds, aes(price, cut)) + 
  ggridges::geom_density_ridges(alpha = 0.5,
                      scale = 1.4,                # ridge height/overlap 
                      rel_min_height = 0.001) +   # where to draw line
  scale_x_continuous(limits = c(0, 22000),        # set cutoff 
                     expand = c(0.01, 0),         # allow expansion
                     label = comma) +       
  scale_y_discrete(expand = c(0.1, 0)) +          # padding around graph 
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ridgeplot-1.png)

---

```r
ggplot(diamonds, aes(price, cut)) + 
  geom_density_ridges(scale = 1.4, rel_min_height = 0.001, alpha = 0.5,
                                quantile_lines = TRUE, quantiles = 2) +  
  scale_x_continuous(limits = c(0, 22000), 
                     expand = c(0.01, 0), label = comma) +       
  scale_y_discrete(expand = c(0.1, 0)) + 
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  theme_diamonds()
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ridgeplotm-1.png)

---

```r
library(viridis)
ggplot(diamonds, aes(price, cut, fill = ..x..)) + 
  geom_density_ridges_gradient(scale = 1.4, rel_min_height = 0.001, 
                               gradient_lwd = 1) +  ## add gradient  
  scale_x_continuous(limits = c(0, 2000), 
                     expand = c(0.01, 0), label = comma) +       
  scale_y_discrete(expand = c(0.1, 0)) + 
  viridis::scale_fill_viridis(name = "Price", option = "D") +
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ridgeplotv-1.png)

---

```r
ggplot(diamonds, aes(price, cut, fill = factor(..quantile..))) + 
  stat_density_ridges(scale = 1.4, rel_min_height = 0.001, 
                      geom = "density_ridges_gradient", calc_ecdf = TRUE, 
                      quantiles = 4, quantile_lines = TRUE) +  
  scale_x_continuous(limits = c(0, 22000), 
                     expand = c(0.01, 0), label = comma) +       
  scale_y_discrete(expand = c(0.1, 0)) + 
  viridis::scale_fill_viridis(discrete = TRUE, name = "Quartiles") +
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ridgeplotvq-1.png)

---

```r
ggplot(diamonds, aes(price, cut)) + 
  geom_density_ridges(scale = 1.4, rel_min_height = 0.001, 
                      alpha = 0.5, size = 0.25, jittered_points = TRUE,  
                      point_shape = "|", point_size = 3, 
                      position = position_jitter(height = 0)) + 
  scale_x_continuous(limits = c(0, 22000), 
                     expand = c(0.01, 0), label = comma) +       
  scale_y_discrete(expand = c(0.1, 0)) + 
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  theme_diamonds() 
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ridgeplotj-1.png)

---

![](ny_hackr_talk_2019_09_19_files/figure-html/raincloud-1.png)

---

```r
## FROM: https://orchid00.github.io/tidy_raincloudplot
source("see code above")

lb <- function(x) mean(x) - ((0.5)*sd(x))
ub <- function(x) mean(x) + ((0.5)*sd(x))

sumld <- diamonds %>% 
  group_by(cut) %>% 
  summarise(mean = mean(price), 
            median = median(price), 
            lower = lb(price), 
            upper = ub(price))

ggplot(data = diamonds, aes(y = price, x = cut, fill = cut)) +
  geom_flat_violin(position = position_nudge(x = .2, y = 0), alpha = .8) +
  geom_point(aes(y = price, color = cut), 
             position = position_jitter(width = .15), 
             size = .5, alpha = 0.8) +
  geom_point(data = sumld, aes(x = cut, y = mean), 
             position = position_nudge(x = 0.3), size = 2.5) +
  geom_errorbar(data = sumld, aes(ymin = lower, ymax = upper, y = mean), 
                position = position_nudge(x = 0.3), width = 0) +
  scale_y_continuous(limits = c(0, 22000), 
                     expand = c(0.01, 0), label = comma) + 
  scale_x_discrete(expand = c(0.1, 0)) +
  labs(x = "Price", y = "Cut", title = "Diamond Prices By Cut") + 
  guides(fill = FALSE) +
  guides(color = FALSE) +
  coord_flip() +
  theme_diamonds() 
```

---

### TidyTuesday

[Github](https://github.com/rfordatascience/tidytuesday), [Shiny app: tidytuesday.rocks](https://nsgrantham.shinyapps.io/tidytuesdayrocks/)

![](images/ttrocks.png)

---

```r
plot <- 
  theme_void() + 
  theme(text = element_text(family = "Gaegu"),
        axis.text = element_text(size = 15, color = "white"),
        axis.title = element_text(size = 20, color = "white"),
        plot.title = element_text(size = 25,face = "bold", 
                                  color = "white", hjust = 0.5),
        plot.caption = element_text(color = "white"),
        plot.margin = margin(t = 1, r = 3, b = 1, l = 1, "cm"),
        axis.title.y = element_text(angle = 90))

ggimage::ggbackground(plot, "chalkboard_simpsons.gif", by = "height")
```

![](ny_hackr_talk_2019_09_19_files/figure-html/simpsons2-1.png)

---

```r
library(magick)
propic <- image_read(here::here("images/propic.png"))
bigdata <- image_read('https://jeroen.github.io/images/bigdata.jpg')
bdl <- image_scale(image_rotate(image_background(propic, "none"), 300), "x150")
ic <- image_composite(image_scale(bigdata, "x400"), bdl, offset = "+170+140")
image_write(ic, here::here("images/ic.png"))
```

![](images/ic.png)

---

```r
library(magick)
jared <- image_read(here::here("images/jared.jpeg"))
j1 <- image_annotate(jared, "hey....
I heard you 
were the 
September speaker 
for the meetup", 
size = 20, 
gravity = "east", color = "white")

image_write(j1, here::here("images/j1.png"))
```

![](images/j1.png)

---

```r
f1 <- image_read(here::here("images/fig.png"))
s1 <- image_read(here::here("images/simpsons.jpeg"))
out <- image_composite(f1, s1, offset = "+1600+30")
image_write(out, here::here("images/out.png"))
```

![](images/out.png)

---

```r
ggplot(GOT_pets, aes(fct_infreq(animals_name), fill = species)) + 
  geom_bar() + 
  scale_fill_got_d(option = "Targaryen", direction = - 1) +
  labs(title = "A Goat Named Arya", 
       subtitle = "Game of Thrones Inspired (Maybe) Pet Names in Seattle", 
       x = element_blank(), y = element_blank(), caption = "@MaraAlexeev") +
  theme_minimal() +
  theme(legend.position = c(0.9, 0.6)) +
  labs(fill = "Pet Species") +
  theme(text = element_text(size = 9, family = "Cinzel"))
```

![](ny_hackr_talk_2019_09_19_files/figure-html/got2-1.png)

---

### Alluvial Plot 
![](ny_hackr_talk_2019_09_19_files/figure-html/myalluvial-1.png)

---

### Alluvial Plot Code

```r
sim_data_pre_post <- read_csv(here::here("sim_data_pre_post.csv")) %>% 
  select(student, unit_title, assessment, score_level) %>% 
  mutate(score_level = factor(score_level), 
         assessment = factor(assessment, levels = c("pre", "post"))) %>% 
  spread(assessment, score_level)  # get wide dataset (pivot_wider)

head(sim_data_pre_post, 5)
```

```
## # A tibble: 5 x 4
##   student unit_title                      pre   post 
##     <dbl> <chr>                           <fct> <fct>
## 1       1 Dinosaur Domestication          2     4    
## 2       1 Earth's Giant Turtle            2     1    
## 3       1 Life on Mars                    1     2    
## 4       1 Potions                         1     4    
## 5       1 Unicorn Traits and Reproduction 1     4
```

---

### Alluvial Plot Code

```r
sim_data_pre_post <- sim_data_pre_post %>% 
  ggalluvial::to_lodes_form(key = "assessment", axes = 3:4) # set up dataset

head(sim_data_pre_post, 5)
```

```
## # A tibble: 5 x 5
##   student unit_title                      alluvium assessment stratum
##     <dbl> <chr>                              <int> <fct>      <fct>  
## 1       1 Dinosaur Domestication                 1 pre        2      
## 2       1 Earth's Giant Turtle                   2 pre        2      
## 3       1 Life on Mars                           3 pre        1      
## 4       1 Potions                                4 pre        1      
## 5       1 Unicorn Traits and Reproduction        5 pre        1
```

---

### Alluvial Plot Code

```r
sim_data_pre_post %>% 
  ggplot(aes(x = assessment,              # categorical x var (pre or post)
             stratum = fct_rev(stratum),  # categorical var (score_level)
             alluvium = alluvium,         # individual/unit (student)
             fill = fct_rev(stratum),     # color of fill
             label = fct_rev(stratum))) +
  ggalluvial::geom_flow(alpha = 0.5) +
  ggalluvial::geom_stratum() +
  scale_fill_manual("Score Level",
                    values = c("1" = "#4d5050", "2" = "#c2c5c6", 
                               "3" = "#f2ac80", "4" = "#F37321")) +
  scale_y_continuous(labels = comma) +
  labs(x = "") +
  facet_wrap(~unit_title, scales = "free_y") + 
  theme_minimal() + 
  theme(panel.border = element_blank(), 
        panel.grid.major = element_blank(),
        panel.grid.minor = element_blank(), 
        axis.line = element_line(colour = "grey"), 
        axis.text.x = element_text(size = 18), 
        axis.text.y = element_text(size = 14), 
        strip.text.x = element_text(size = 16), 
        legend.position = "bottom", 
        legend.title = element_text(size = 16),
        legend.text = element_text(size = 16))
```

---

![](ny_hackr_talk_2019_09_19_files/figure-html/ggalluvial2-1.png)

---

```r
## FROM: https://mdneuzerling.com/post/my-data-science-job-hunt/

job_outcomes %>% 
  mutate(final_outcome = coalesce(outcome, `2nd stage`, `1st stage`)) %>% 
  ggalluvial::to_lodes_form(key = "contact", axes = 2:5) %>% 
  ggplot(aes(x = contact, stratum = stratum, 
             alluvium = alluvium, label = stratum)) + 
  geom_alluvium(aes(fill = final_outcome), 
                color = "darkgrey", na.rm = TRUE) +
  geom_stratum(na.rm = TRUE) +
  geom_text(stat = "stratum", na.rm = TRUE, size = 1) + 
  scale_fill_manual(values = c("ghosted" = "#F0E442", 
                               "no role" = "#CC79A7",
                               "withdrew" = "#0072B2", 
                               "rejected" = "#D55E00",
                               "offer" = "#009E73")) + 
  labs(x = "", fill = "Final Outcome", 
       caption = "David Neuzerling @mdneuzerling") + 
  theme_minimal(legend.position = "bottom") + 
  theme(text = element_text(size = 12)) 
```

---

![](ny_hackr_talk_2019_09_19_files/figure-html/annotation-1.png)

---

```r
plot <- ggplot(df_scores, aes(x = reorder(publisher, average_score), 
                              y = metascore)) +
  geom_jitter(aes(color = publisher), size = 5, alpha = 0.25, width = 0.15) +
  ggforce::geom_mark_circle(x = 10, y = 94, color = 'grey50', 
                            label.fill = NA, expand = unit(4, "mm")) +
  geom_segment(aes(x = publisher, xend = publisher, 
                   y = total_avg, yend = average_score), 
               size = 0.5, color='gray30') +
  geom_point(mapping = aes(x = publisher, y = average_score, fill = publisher), 
             color = "gray30", shape = 21, size = 7, stroke = 1) +
  geom_hline(aes(yintercept = total_avg), color = "gray30", size = 0.5) +
  annotate("text", x = 6.6, y = 86, fontface = "bold", 
           label = 'Average Overall') +
  annotate("text", x = 6.3, y = 86, 
           label = glue::glue('{round(total_avg, 1)} Metascore')) +
  annotate("text", x = 2.5, y = 55, fontface = "italic", 
           label = 'Average per publisher') +
  annotate("text", x = 9.7, y = 45, fontface = "bold", 
           label = 'Worst Game Overall') +
  annotate("text", x = 9.4, y = 45, 
           label = "Rogue Warrior") +
  annotate("text", x = 9.6, y = 88, fontface = "bold", 
           label = 'Best Games Overall') +
  annotate("text", x = 9.3, y = 88, 
           label = "Elder Scroll Series") +
  coord_flip() +
  scale_y_continuous(limits = c(25, 100))+
  theme_minimal() +
  guides(color = FALSE, fill = FALSE) +
  labs(title = "PC Game Ratings by Publisher",
       subtitle = "Ratings based on Metascore out of 100",
       caption = "Tanya Shapiro @tanya_shapiro",
       x = "", y = 'METASCORE', color = "# of Owners")
```

---

```r
arrows <- tibble(
  x1 = c(6.2,        #Avg Overall
         2.6,        #Avg Per Publisher
         2.6,        #Avg Per Publisher
         9.4,        #Worst Game
         9.7),       #Best Game
  x2 = c(5.5,        #Avg Overall
         4.1,
         3.1,
         9.9,
         9.9),
  y1 = c(86,         #Avg Overall
         55,         #Avg Per Publisher
         55,         #Avg Per Publisher
         40,
         88),
  y2 = c(total_avg,  #Avg Overall
         70.2,       #Avg Per Publisher
         68.2,       #Avg Per Publisher
         30,
         92))

# add arrows
p <- 
  plot + 
  geom_curve(data = arrows, aes(x = x1, y = y1, xend = x2, yend = y2),
             arrow = arrow(length = unit(0.07, "inch")), size = 0.6,
             color = "gray20", curvature = -0.25)
```

---

[Cédric Scherer, king of #tidytuesday](https://github.com/Z3tt/TidyTuesday)

![](images/2019_18_BirdCollisions.png)

![](images/2019_17_animes.png)

---

```r
## FROM: https://twitter.com/lpmkremer/status/1168912680767447040?s=21

dat <- bind_rows(
  tibble(x = rnorm(7000, sd = 1), y = rnorm(7000, sd = 10), group = "foo"),
  tibble(x = rnorm(3000, mean = 1, sd = .5), 
         y = rnorm(3000, mean = 7, sd = 5), group = "bar"))

ggplot(data = dat, mapping = aes(x = x, y = y)) +
  geom_pointdensity(adjust = 3) +
  scale_color_viridis() + 
  theme_minimal()
```

![](ny_hackr_talk_2019_09_19_files/figure-html/ggpointdensity-1.png)

---

# In Summary

- Change your background

- Pick a nice font and font size

- Annotate away!

- Use ggplot and ggplot adjacent packages

- Add images

- Follow #TidyTuesday

- HAVE FUN!

![](images/pie.jpeg)

---

### More resources

Take a sad plot and make it better (from Alison Hill: 
  https://alison.rbind.io/talk/2018-ohsu-sad-plot-better/
  
BBC Visual and Data Journalism cookbook for R graphics: 
  https://bbc.github.io/rcookbook/

The Economist: 
  https://medium.economist.com/mistakes-weve-drawn-a-few-8cdd8a42d368
  
New way to make alluvial charts: 
  https://ggforce.data-imaginist.com/reference/geom_parallel_sets.html
  
You Can Design a Good Chart with R (on the Tufte tweetstorm): 
  https://medium.com/nightingale/you-can-design-a-good-chart-with-r-5d00ed7dd18e
  
ggridges gallery: https://cran.r-project.org/web/packages/ggridges/vignettes/gallery.html
  
  
---

# Thank you

- Samuel Crane, @samuelcrane and Amplify

- Sebastian Teran Hidalgo, @steranhidalgo

- RLadies-NYC, @RLadiesNYC

- Jared Lander, @jaredlander

- all the #TidyTuesday contributors:

Gil Henriques, @_Gil_Henriques

Mara Alexeev, @MaraAlexeev

David Neuzerling, @mdneuzerling

Tanya Shapiro, @tanya_shapiro

Cédric Scherer, @CedScherer

- the Rstats community

---