Challenge 5 Solutions

Introduction to Visualization

library(tidyverse)
library(ggplot2)

knitr::opts_chunk$set(echo = TRUE, warning=FALSE, message=FALSE)

Read in data:

cereal <- read_csv("_data/cereal.csv")

The “cereal” data set includes brands of cereal with their corresponding sugar and sodium content. The data set also includes a variable titled “category” with values of “A,” “B,” and “C,” but it is unclear what these values refer to. The data set is clean as-is and does not require mutation, so I will move on to data visualization.

Univariate visualization:

I will first generate a histogram to illustrate the sugar content of cereals so we can see the distribution of values across the count of cereal types.

# histogram:

# cereal count by sugar content:

ggplot(cereal, aes(x=Sugar)) + 
  geom_histogram(binwidth=1, fill="lightpink", color="white", alpha=0.9)

I will next generate a density plot. Given that this is a univariate visualization to capture sodium content by the count of cereal types, this chart won’t give us much information; regardless, I wanted to try it! I will add a mean indicator to give us a bit more insight, which shows the average sodium content to be approximately 260 (I assume) miligrams.

# density plot:

# cereal count by sodium content:

cereal %>%
  ggplot( aes(x=Sodium)) +
    geom_density(fill="plum3", color="white", alpha=0.9) +
  geom_vline(aes(xintercept=mean(Sodium)),
            color="white", linetype="dashed", size=1)

I will next generate a boxplot to visualize the distribution of sugar content in a different format.

# try boxplot:

ggplot(cereal, aes(y = Sugar)) +
    geom_boxplot() +
  geom_boxplot(fill="cyan3", color="black", alpha=0.9)

Bivariate visualization:

Next, I will generate a dot plot capturing sodium content by cereal brand. Because this is a relatively small data set, I thought I’d try using a dot plot instead of a histogram.

# dot plot: 

library(lattice)

dotplot(cereal$Cereal ~ cereal$Sodium)

Next, I chose to create a bivariate visualization capturing sodium content by sugar content. I also added value labels to each data point so we can see the cereal brand in reference.

# scatter plot:

library(ggrepel)

ggplot(cereal, aes(y=Sugar, x=Sodium)) +
  geom_point() +
  geom_text_repel(aes(label = Cereal))