Beer is a beverage with a rich diversity of styles, each boasting its own unique history and characteristics. The American craft beer market offers a wide array of these styles, raising intriguing questions: What distinguishes these different styles of beer? Which states have a rich craft beer culture? This project aims to explore these varied beer styles from multiple perspectives.
This project utilizes a dataset from Kaggle.com, featuring 2,410 American craft beers and 510 breweries, gathered from CraftCans.com in January 2017. The dataset includes a range of variables, but for this visualization, we concentrate on those pertinent to our theme: beer style, ABV (Alcohol By Volume), IBU (International Bitterness Units), and the state of origin.
The visualization process starts with data analysis conducted using the R programming language. The interactive visualizations are created using D3.js, a JavaScript library specializing in data-driven visualizations. Non-interactive visuals are developed with ggplot2 in R and then refined in vector design software, Sketch, for enhanced visual appeal.
In the upcoming visualizations, we explore the variety of beers produced in the United States in 2017, focusing specifically on different beer styles. This exploration includes an in-depth comparison of the five most prevalent styles, analyzing them through the lenses of taste profile and geographical distribution.
How many different styles of craft beer are available in the market? What are the most common styles?
What are the differences in bitterness and alcohol content among different beer styles?
Where in the USA are the most common types of beer produced?
The treemap visualization displays various beer styles, with each rectangle's area indicating the market presence of that style, measured by the number of products. Alongside, I include the percentage of each style* to highlight quantitative differences. This treemap is chosen for its dual ability to show the variety of beer styles and their respective market shares, effectively aligning with my goal to showcase the diversity of the market and pinpoint the most common styles for further analysis. *The percentage of each style is calculated by dividing the number of products in that style by the total number of products.
Beer product entries were grouped by style. To enhance the treemap's readability, styles with fewer entries were excluded. The cutoff was set at 30 products; styles with fewer were were considered as less relevant. This process identified five styles with the most products, which were then used as a focus group in subsequent visualizations.
Dive into the bitterness landscape of different beer styles with the charts below. Each line chart maps out the number of beer products within various bitterness levels for a selected style. The x-axis measures the bitterness in IBU – the higher the number, the bitterer the beer. The y-axis counts the beers that match each bitterness score. Look for peaks in the lines; these highlight the most common levels of bitterness within each style. Just click on a style to start comparing!
The International Bitterness Unit (IBU) measures beer bitterness. A lower IBU means less bitterness, while a higher number indicates more bitterness. It's important to remember that IBU is not the sole determinant of beer's bitterness – malt and other flavors also influence the perceived taste.
HOVER OVER DOTS TO SEE DETAIL DATA
CLICK TO VIEW DIFFERENT STYLES
A set of line charts enables the comparison of the bitterness of the five most common beer styles. By clicking the "Style" button, viewers can compare the bitterness distributions across these five styles. Consistent color coding for 5 beer style throughout the project can enhance overall readability.
Given the vast array of beer types, this visualization focuses on the five most common styles identified in the previous section. Using IBU and product frequency as variables, a single line chart is drawn for each bear style to show the distribution of products across different bitterness levels.
Curious about the strength of different beer styles? See side-by-side comparisons of the top five beer styles by alcohol content. The x-axis marks the alcohol percentage or ABV, Higher numbers on the x-axis indicate stronger beers. The y-axis shows the number of products. Peaks highlight the common ABV levels.
Alcohol by Volume (ABV) indicates the percentage of alcohol in a beverage. It's calculated by the volume of ethanol in 100 ml of the total liquid. For instance, a 2% alcohol content equates to 0.02 ABV.
HOVER OVER DOTS TO SEE DETAIL DATA
CLICK TO VIEW DIFFERENT STYLES
Explore the relationship between alcohol content (ABV) and bitterness (IBU) across beer styles with our interactive scatter plot. The x-axis shows IBU for bitterness, and the y-axis shows ABV for alcohol strength, with colors differentiating the beer styles. Click to filter by style and see the correlation.
HOVER OVER DOTS TO SEE DETAIL DATA
The scatter plot simultaneously displays the ABV and IBU, revealing the interplay between alcohol strength and bitterness. Each dot represents an individual beer product, with a total of 667 plotted, allowing for a detailed examination of each beer's characteristics compared to others. Typically, a higher ABV is associated with increased bitterness.
This circular grouped bar chart showcases the production of the five most common beer styles across various states in the U.S. It highlights that Colorado and California are leading producers, with Colorado being the top for four styles.
Above is a grouped bar chart with a circular layout. The data is divided into five groups based on five beer styles, showing the number of beer products produced in different states. From the figure, we can see that Colorado and California produce more beer. Especially Colorado, it is the largest producer of the four styles.
Colors are used to encode different styles. Within each group, bars are arranged in descending order, highlighting the states with higher production. The circular layout optimizes space but can make data reading challenging. To address this, prominent auxiliary lines, or circular grids, are overlaid on the bars for enhanced readability.
To create a more readable visualization with reduced visual and information clutter, I focused on states with higher production by removing those with less than five products from the dataset.