Designing effective visualizations

Intro to Data Analytics

Designing effective visualizations

Keep it simple

Use color to draw attention

• Plot
• Code

d %>%
  mutate(category = str_replace(category, " ", "\n")) %>%
  ggplot(aes(x = category, y = value, fill = category)) +
  geom_col() +
  theme_minimal() +
  labs(x = "", y = "") +
  theme(legend.position = "none")

Use color to draw attention

• Plot
• Code

ggplot(d, aes(x = fct_reorder(category, value), y = value, fill = category)) +
  geom_col() +
  theme_minimal() +
  coord_flip() +
  labs(x = "", y = "") +
  scale_fill_manual(values = c("red", rep("gray", 4))) +
  theme(legend.position = "none")

Tell a story

Principles for effective visualizations

Principles for effective visualizations

• Order matters
• Put long categories on the y-axis
• Keep scales consistent
• Select meaningful colors
• Use meaningful and nonredundant labels

Data

In September 2019, YouGov survey asked 1,639 GB adults the following question:

In hindsight, do you think Britain was right/wrong to vote to leave EU?

• Right to leave
• Wrong to leave
• Don’t know

Source: YouGov Survey Results, retrieved Oct 7, 2019

Order matters

Alphabetical order is rarely ideal

• Plot
• Code

ggplot(brexit, aes(x = opinion)) +
  geom_bar()

Order by frequency

• Plot
• Code

fct_infreq: Reorder factors’ levels by frequency

ggplot(brexit, aes(x = fct_infreq(opinion))) + 
  geom_bar()

Wrap the opinion variable inside the fct_infreq function and reorder the categories by frequency

Clean up labels

• Plot
• Code

ggplot(brexit, aes(x = fct_infreq(opinion))) +
  geom_bar() +
  labs( 
    x = "Opinion", 
    y = "Count" 
  ) 

Alphabetical order is rarely ideal

• Plot
• Code

ggplot(brexit, aes(x = region)) +
  geom_bar()

Use inherent level order

• Relevel
• Plot

fct_relevel: Reorder factor levels using a custom order

brexit <- brexit %>%
  mutate(
    region = fct_relevel( 
      region,
      "london", "rest_of_south", "midlands_wales", "north", "scot"
    )
  )

Clean up labels

• Recode
• Plot

fct_recode: Change factor levels by hand

brexit <- brexit %>%
  mutate(
    region = fct_recode( 
      region,
      London = "london",
      `Rest of South` = "rest_of_south",
      `Midlands / Wales` = "midlands_wales",
      North = "north",
      Scotland = "scot"
    )
  )

Put long categories on the y-axis

Long categories can be hard to read

Move them to the y-axis

• Plot
• Code

ggplot(brexit, aes(y = region)) + 
  geom_bar()

And reverse the order of levels

• Plot
• Code

fct_rev: Reverse order of factor levels

ggplot(brexit, aes(y = fct_rev(region))) + 
  geom_bar()

Clean up labels

• Plot
• Code

ggplot(brexit, aes(y = fct_rev(region))) +
  geom_bar() +
  labs( 
    x = "Count", 
    y = "Region" 
  ) 

Pick a purpose

Segmented bar plots can be hard to read

• Plot
• Code

ggplot(brexit, aes(y = region, fill = opinion)) + 
  geom_bar()

Use facets

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = region)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1) 

Avoid redundancy?

ggplot(brexit, aes(y = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1)

Redundancy can help tell a story

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1)

Be selective with redundancy

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1) +
  guides(fill = "none") 

Use informative labels

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1) +
  guides(fill = "none") +
  labs(
    title = "Was Britain right/wrong to vote to leave EU?", 
    x = NULL, y = NULL
  )

A bit more info

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1) +
  guides(fill = "none") +
  labs(
    title = "Was Britain right/wrong to vote to leave EU?",
    subtitle = "YouGov Survey Results, 2-3 September 2019", 
    caption = "Source: https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/x0msmggx08/YouGov%20-%20Brexit%20and%202019%20election.pdf", 
    x = NULL, y = NULL
  )

Add a source for our visualization

Let’s do better

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1) +
  guides(fill = "none") +
  labs(
    title = "Was Britain right/wrong to vote to leave EU?",
    subtitle = "YouGov Survey Results, 2-3 September 2019",
    caption = "Source: bit.ly/2lCJZVg", 
    x = NULL, y = NULL
  )

Shorten the URL using bit.ly

Fix up facet labels

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region,
    nrow = 1,
    labeller = label_wrap_gen(width = 12) 
  ) + 
  guides(fill = "none") +
  labs(
    title = "Was Britain right/wrong to vote to leave EU?",
    subtitle = "YouGov Survey Results, 2-3 September 2019",
    caption = "Source: bit.ly/2lCJZVg",
    x = NULL, y = NULL
  )

Add the labeller argument to the facet_wrap function

Select meaningful colors

Rainbow colors not always the right choice

Manually choose colors when needed

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1, labeller = label_wrap_gen(width = 12)) +
  guides(fill = "none") +
  labs(title = "Was Britain right/wrong to vote to leave EU?",
       subtitle = "YouGov Survey Results, 2-3 September 2019",
       caption = "Source: bit.ly/2lCJZVg",
       x = NULL, y = NULL) +
  scale_fill_manual(values = c( 
    "Wrong" = "red", 
    "Right" = "green", 
    "Don't know" = "gray" 
  )) 

Choosing better colors

colorbrewer2.org

Use better colors

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1, labeller = label_wrap_gen(width = 12)) +
  guides(fill = "none") +
  labs(title = "Was Britain right/wrong to vote to leave EU?",
       subtitle = "YouGov Survey Results, 2-3 September 2019",
       caption = "Source: bit.ly/2lCJZVg",
       x = NULL, y = NULL) +
  scale_fill_manual(values = c(
    "Wrong" = "#ef8a62", 
    "Right" = "#67a9cf", 
    "Don't know" = "gray" 
  ))

Select theme

ggplot complete themes guide

• Plot
• Code

ggplot(brexit, aes(y = opinion, fill = opinion)) +
  geom_bar() +
  facet_wrap(~region, nrow = 1, labeller = label_wrap_gen(width = 12)) +
  guides(fill = "none") +
  labs(title = "Was Britain right/wrong to vote to leave EU?",
       subtitle = "YouGov Survey Results, 2-3 September 2019",
       caption = "Source: bit.ly/2lCJZVg",
       x = NULL, y = NULL) +
  scale_fill_manual(values = c("Wrong" = "#ef8a62",
                               "Right" = "#67a9cf",
                               "Don't know" = "gray")) +
  theme_minimal()