Choosing a chart

Five intentions (compare, trend, composition, distribution, relationship) and the chart for each, how precisely visual encodings can be read, and four traps: truncated axis, dual axis, 3D effect, too many categories.

  • 24 cards
  • 6 topics
  • Free

The outline, topic by topic

Each topic is weighted the way the exam weighs it.

  1. Concepts
  2. Questions
  3. Sort and link
  4. The numbers
  5. Scenarios
  6. Understand

A few cards from the deck

  • Data visualisation: what is it?

    Show the answer

    A representation of data through shapes (position, length, colour) so the eye can quickly spot a comparison, a trend or an exception.

    The choice starts from the question asked of the data: the same table of figures calls for different charts depending on whether you compare, track or look for a link.

  • Which chart shows the share of each of the three channels (shop, website, phone) in the year's revenue?

    • A pie chart with three slices
    • A histogram with three bins, one per channel
    • A line joining the three channels
    • A scatter plot of channel against revenue
    Show the answer

    A pie chart with three slices

    Showing each element's share of a whole is a composition: a pie fits when the slices are few (so does a 100% stacked bar). A histogram splits a continuous measure into bins, not categories.

  • To follow a trend over time, choose a …; to see how values are distributed (for example customers' ages), a ….

    Show the answer

    To follow a trend over time, choose a line chart; to see how values are distributed (for example customers' ages), a histogram.

    The line joins points ordered in time and shows the slope. The histogram groups values into classes and shows the shape of their spread.

  • Chart type: what is it?

    Show the answer

    A standard form of representation, such as bars, line, scatter plot, histogram or pie, suited to an intention of reading.

    Bars: compare. Line: follow over time. Scatter plot: relate two measures. Histogram: see a distribution. Pie: a composition, with few slices.