Chart selection, visual encodings, color use, Tufte principles, dashboards, misleading charts and storytelling with data.
35 cards · basic cards · AI-written, checked twice. Edit anything.
- When should you use a bar chart instead of other chart types?
Show answer
For comparing values across categories, especially when the number of categories is moderate (up to about 20).- When is a line chart most appropriate?
Show answer
For showing changes over time, especially for continuous time series or when you want to emphasize trends.- What is a scatterplot best used for?
Show answer
Showing the relationship between two continuous variables and identifying correlations or clusters.- When is a pie chart an appropriate choice?
Show answer
Only when showing parts of a whole that sum to 100%, and only with 2-3 categories since humans struggle comparing angles.- What is a heatmap used for?
Show answer
Visualizing values in a matrix where color intensity represents the magnitude, useful for showing patterns in high-dimensional data.- When should you use a box plot?
Show answer
For displaying the distribution of continuous data and comparing distributions across groups, showing quartiles and outliers.- What is a small multiple visualization?
Show answer
A series of similar charts showing the same data structure but for different subsets or categories, enabling easy comparison.- When are parallel coordinates plots useful?
Show answer
For exploring relationships in high-dimensional datasets (many variables) by drawing each observation as a line across parallel axes.- What are the six main visual encodings in order of effectiveness?
Show answer
Position, length, direction, angle, area, and color.- What is preattentive processing in visualization?
Show answer
The ability to perceive certain visual properties (like color or position) in under 250 milliseconds without conscious effort.- Why is position the most effective visual encoding?
Show answer
Because humans are naturally skilled at judging spatial position and distance along a common scale.- How should quantitative data be encoded in a visualization?
Show answer
Using position (on a common scale), length, or area rather than color, angle, or size when possible.- How should categorical data be encoded?
Show answer
Using position, color, or shape, rather than position on a continuous scale which implies magnitude.- What is the Gestalt principle of proximity in visualization?
Show answer
Objects that are close together are perceived as a group, helping organize visual elements.- What are the three main types of color palettes?
Show answer
Sequential (light to dark for ordered data), diverging (two hues from a midpoint), and categorical (distinct hues for unordered categories).
FlashKeepers