Home Up PDF Prof. Dr. Ingo Claßen
Visualization - Amounts - DSML

Attribution

The following slides are based on

  • Fundamentals of Data Visualization (link)
    and are following the
  • Attribution-NonCommercial-NoDerivatives 4.0 International License (link)

Bar Charts

Magnitudes of amounts

Comparison of amounts

Quantitative values per categorie

Values must start at zero

Common problem: labels too long

Rotated Label

Not optimal

Actually, matter of taste

Horizontal Bar Charts

Better solution in case of long labels

Natural Order on Labels

In this case the sorting according to labels is sensible

Sorting according to values would prevent insights

Grouped Barchars

Amounts for more than one Category

Demographics per Age Bands

Age Bands per Demographics

Grouped Bar Plots

Here Age Bands per Demographics

No color coding necessary

Stacked Barchars

Totals for items of one category recognizable

Here per cabin class

Wrong Usage of Bar Chart

Bar chart that doesn't start at 0

Biased presentation

Differences seem to be greater than they are

Start at 0

Unbiased Presentation

Dot Plots

Barcharts must start at zero

Might hinder comprehension

Might disguise diferences. Differences seem to be small

But small differences in life expectancy might be important

Dot plots must not start at zero

Heatmaps

Large number of categories/combinations

Barchart not suitable

Here sorting according to amount in 2016