Judgments of change and proportion in graphical perception.

Area of Science:

• Data visualization
• Cognitive psychology
• Human-computer interaction

Background:

• Effective graphical displays are crucial for data interpretation.
• Understanding how different graph types affect perception is essential for clear communication.
• Previous research indicates variations in how humans process visual data.

Purpose of the Study:

• To compare the accuracy and speed of judging data change and proportion across different graph types.
• To investigate the influence of graph design on perceptual efficiency.
• To provide evidence-based recommendations for optimizing graphical display design.

Main Methods:

• Two experiments were conducted where subjects judged change and proportion from various graphs (line, bar, pie, tiered bar, divided bar).
• Performance metrics included judgment speed and accuracy.
• The presence or absence of a graduated scale was a key variable.

Main Results:

• Line and bar graphs facilitated quicker and more accurate judgments of change compared to pie or tiered bar graphs.
• This difference was more pronounced with smaller rates of change.
• Pie charts and divided bar graphs enabled faster and more accurate proportion judgments, especially without a graduated scale.
• Results align with the proximity compatibility principle.

Conclusions:

• Perception of change is more direct with line and bar graphs.
• Perception of proportion is more direct with pie charts and divided bar graphs.
• These findings offer insights for designing more effective and intuitive graphical displays.