Categorical Data Visualizations F

What is this?

What is this?

simple bar chart

What is this?

simple pie charts

What is this?

stacked bar chart

What is this?

grouped bar chart

What is this?

mosaic plot

What is this?

sankey diagram

Definition of a simple bar chart

Helps visualize the magnitude for a single set of categories

Features of a simple pie chart

-circle split into slices that represent values within a category

-slices are sized accordingly to the proportion out of the whole the value represents

-numbers are used to label each slice

-each of the values are distinct to the case

When to use a stacked bar chart?

Useful when the sum of the bars represents a meaningful amount or count. Also nice for illustrating the relative sizes in a given category

When to use a bar chart?

-when there are a lot of values within a category

-the proportions in the categories are similar

-we want to compare the proportions of a category across multiple sets

When to use a grouped bar chart?

Easier to see differences between categories. Use when emphasizing differences

When to use Mosaic Plots?

Helps show relationships and provide a visual way to compare independence between groups.

Definition of Sankey Diagrams

Illustrates the “flow” from one set of categories to another

Marginal Probability

Has to do with one**;** ex, the probability of randomly selecting a passenger from first class or survival status overall

Joint Probability

Multiple (2 max); ex, the probability that a randomly selected passenger survived and was first class

What graph to use in joint probability

grouped bar chart

What graph to use in marginal probability

pie chart

Conditional Probability

the total amount (something like only first class); ex, the probability that a passenger survived given that they were in first class

What graph to use in conditional probability

stacked bar chart

How to asses independence between two categorical variable via a stacked bar chart or mosaic plot?

Putting two bars next to each other that are different categories; the areas are proportionals to the joint probability of each category