[data analysis and visualization] key points of data drawing 12- importance of chart notes

Key points of data drawing 12- Importance of chart notes

Chart annotation is a key component of data visualization . It can turn boring graphics into an interesting and insightful way to convey information . At present, we pay less than five seconds' attention to reading on the Internet . therefore , If you can't get someone's attention in five seconds , Then you may have lost the audience . Adding accurate annotations can greatly help attract the attention of the audience . Use keywords 、 shape 、 Colors and other visual effects to help them get to the point . Annotation is a general concept , There are hundreds of different chart annotation methods , The specific use depends on the actual content , Let's look at a few examples .

Example 1

The main danger in drawing a line chart is that you end up with a spaghetti chart ： Too many groups can often lead to chart confusion , It is difficult to read any valid information , As shown in the figure below .

# Libraries
library(tidyverse)
library(hrbrthemes)
library(babynames)
library(viridis)

#  Reading data 
data <- babynames %>% 
  filter(name %in% c("Mary","Emma", "Ida", "Ashley", "Amanda", "Jessica",    "Patricia", "Linda", "Deborah",   "Dorothy", "Betty", "Helen")) %>%
  filter(sex=="F")

#  mapping 
data %>%
  ggplot( aes(x=year, y=n, group=name, color=name)) +
    geom_line() +
    scale_color_viridis(discrete = TRUE) +
    theme(
      legend.position="none",
      plot.title = element_text(size=14)
    ) +
    ggtitle("A spaghetti chart of baby names popularity")

You may want to highlight your main points , Instead of presenting this graphic to your audience . Then it is more meaningful to show only one variable data ：

data %>%
  mutate( highlight=ifelse(name=="Amanda", "Amanda", "Other")) %>%
  # color Set specific data 
  ggplot( aes(x=year, y=n, group=name, color=highlight, size=highlight)) +
    geom_line() +
    scale_color_manual(values = c("#69b3a2", "lightgrey")) +
    scale_size_manual(values=c(1.5,0.2)) +
    theme(legend.position="none") +
    ggtitle("Popularity of American names in the previous 30 years") +
    geom_label( x=1990, y=55000, label="Amanda reached 3550\nbabies in 1970", size=4, color="#69b3a2") +
    theme(
      legend.position="none",
      plot.title = element_text(size=14)
)

Example 2

When you show the audience several figures . The report consists of the following parts ：1 individual barplot The figure shows the total amount of the company's five products ,1 A line chart shows the evolution of the benefits generated by these five products over the past decade ：

#  Read Library 
library(tidyverse)
library(hrbrthemes)
library(patchwork)
library(viridis)

#  Create data 
data <- data.frame(
  product = LETTERS[1:5],
  value = c(10, 14, 18, 6, 15)
)

#  mapping 
a <- data %>%
  ggplot( aes(x=product, y=value, fill=product)) +
    geom_bar(stat="identity", width=0.5) +
    scale_fill_viridis(discrete=TRUE) +
    ggtitle("How much the products cost") +
    theme(
      plot.title = element_text(size=12),
      legend.position = "none"
    ) +
    xlab("") +
    ylab("Cost in M$")

#  Create data 
data <- data.frame(
  year = rep( seq(1,10), 5 ),
  product = rep(LETTERS[1:5], each=10 ),
  value = c( 
    seq(1,10) + sample( 1:3, 10, replace = TRUE), 
    seq(5,14) + sample( 4:10, 10, replace=TRUE), 
    seq(10,1)*2 + sample( 3:5, 10, replace=TRUE), 
    seq(20,11) + sample( 12:17, 10, replace=TRUE), 
    seq(1,10) + sample( 40:10, 10))
)

#  mapping 
b <- data %>%
  ggplot( aes(x=year, y=value, color=product)) +
    geom_line() +
    scale_color_discrete() +
    scale_x_continuous(breaks=c(1:10)) +
    ggtitle("How much the products make") +
    theme(
      plot.title = element_text(size=12)
    ) +
    xlab("Year") +
    ylab("Benefit in M$")

# Show
a + b

You can see , It's hard to make a connection between two charts , Because each chart uses a different palette . product A Dark purple in the figure on the left , Red in the figure on the right . So when you have multiple shapes , Use the same color for the same item throughout the report .

data <- data.frame(
  product = LETTERS[1:5],
  value = c(10, 14, 18, 6, 15)
)

a <- data %>%
  ggplot( aes(x=product, y=value, fill=product)) +
    geom_bar(stat="identity", width=0.5) +
    scale_fill_viridis(discrete=TRUE) +
    ggtitle("How much the products cost") +
    theme(
      plot.title = element_text(size=12),
      legend.position = "none"
    ) +
    xlab("") +
    ylab("Cost in M$")

data <- data.frame(
  year = rep( seq(1,10), 5 ),
  product = rep(LETTERS[1:5], each=10 ),
  value = c( 
    seq(1,10) + sample( 1:3, 10, replace = TRUE), 
    seq(5,14) + sample( 4:10, 10, replace=TRUE), 
    seq(10,1)*2 + sample( 3:5, 10, replace=TRUE), 
    seq(20,11) + sample( 12:17, 10, replace=TRUE), 
    seq(1,10) + sample( 40:10, 10))
)

b <- data %>%
  ggplot( aes(x=year, y=value, color=product)) +
    geom_line() +
    scale_color_viridis(discrete=TRUE) +
    scale_x_continuous(breaks=c(1:10)) +
    ggtitle("How much the products make") +
    theme(
      plot.title = element_text(size=12)
    ) +
    xlab("Year") +
    ylab("Benefit in M$")

# Show
a + b

Example 3

The ratio of the width to the height of a figure . It can have a significant impact on the insights gained from your graphics .

The following figure is an example of a line chart , It shows the data from 1749 Year to 1983 The annual number of sunspots in . This number is quite obvious in the solar cycle . The number of sunspots follows this pattern of solar activity 11 Annual periodic change , Make this line fluctuate up and down regularly . Obviously , Some peaks are higher than others , The maximum number of smooth sunspots is 50 To 180 Between .

library(tidyverse)
library(hrbrthemes)

#  Reading data 
data <- data.frame(
  Year = as.numeric(time(sunspot.year)),
  Sunspots = as.numeric(sunspot.year)
)

#  mapping 
data %>%
  ggplot( aes(x=Year, y=Sunspots)) +
    geom_line() +
    ggtitle("Number of sunspots per year") +
    theme(
      plot.title = element_text(size=12)
    )

Now let's change the aspect ratio of this graph , Make it wider and lower . To exaggerate the effect , Enlarge the drawing to 1800 Year to 1900 year . Here you can see the new results . The number of sunspots increases faster than it decreases . So try several different aspect ratios . This exploration greatly improves the value of graphics ：

data %>%
  filter(Year>1800 & Year<1900) %>%
  ggplot( aes(x=Year, y=Sunspots)) +
    geom_line() +
    ggtitle("Number of sunspots per year") +
    theme(
      plot.title = element_text(size=12)
    )

Reference resources

• ANNOTATION IS CRUCIAL FOR YOUR DATAVIZ
• MIND THE ASPECT RATIO
• MENTAL ARITHMETIC IN DATAVIZ