Pluralsight Logo
Author avatar

Hemant Singh

Author badge Author

Comparing Visualization Types for Your Data with Proportions

Hemant Singh

Author BadgeAuthor
  • Nov 8, 2018
  • 4 Min read
  • Nov 8, 2018
  • 4 Min read
Data Visualization

Nightingale Rose Chart

  • This is also known as Coxcomb Chart or Polar Area Diagram
  • Use: It is used to plot cyclic phenomenon with segments
  • It is a combination of stacked column chart and radar chart
  • The chart is similar to the pie chart, but the angles are equal and the amount in which radial regions extends from the center of the circle
  • The radius of each sector is proportional to the square root of the value of the variable, so the area of each sector represents the value of the variable
  • It was famously used by Florence Nightingale to represent preventable deaths, death from wounds, and death by all other causes of soldiers during the war
  • The outer regions have larger area size, so they get more emphasis by the viewer. This is a con of this visualization
  • Purpose Analyzing cyclic phenomenon and contribution of different factors using segments Identifying the relative effects across different periods visually * Identifying patterns if any in the cyclin phenomenon

Dot Matrix Chart

  • Use: Gives a quick overview of proportions of categories in a dataset
  • Dots are used in the matrix using different colors for each category
  • The number of dots is proportional to the size or value of the category
  • Another version of the dot matrix is where icons are used instead of dots in the matrix and they are grouped by similar color
  • If a single variable is in the dataset, then dot matrix chart can be used to display proportions across different datasets
  • Purpose Compare distribution proportion across different datasets A very simple view of proportions to discover patterns

Circle Packing

  • It is also known as a circular tree map
  • Use: Visualize hierarchic organization
  • It is a variation of tree map and each node is represented by a circle and sub-nodes are represented as circles inside it
  • The size of each circle is proportional to the value
  • Colors are used to represent categories or groupings across or inside the hierarchy
  • It is difficult for the viewer to compare exact size when it comes to circles, but it gives a very good, easy way to consume and neat representation of hierarchy and groupings
  • Purpose To compare the values of a group precisely To represent organizations of groups into subgroups

Proportional Area Chart

  • Use: Display proportions for quick view without using scales
  • Shapes such as circles, rectangles are given area proportional to the value being represented
  • It is difficult to gauge exact value because of lack of scale
  • Purpose: To get a quick idea of the proportional distribution

Stacked Bar Graph

  • Use: Display segments of a dataset in a bar chart and contribution of each segment to the total
  • The bars in the chart represent as the whole and the segments of bars represent different parts or categories of the whole
  • These graphs should be used when the sum of the values is as important as the individual contributors
  • The bars of the graph can be horizontal or vertical
  • Stacked bar graphs can have one category axis which describes the categories and up to two numerical axes which can be two scales for the data
  • If the number of segments are very high, these charts become harder to read
  • Purpose Finding relationship of each part to different categories Displaying distribution of variables in different categories * Comparing part by part role of one variable across different categories