Angle judgments are markedly less accurate than length under equal conditions

What it is

Task-conditioned angle versus length judgment compares the two encodings within the same judgment task. The title summarizes a common design warning, not a rule stably established by matched experiments. Cleveland and McGill's theoretical ranking predicted angle to be worse than length, but their original angle and length experiments used different formats and cannot be directly compared; Heer and Bostock's matched task did not support that prediction. The safer conclusion is that common-scale position generally outperforms both, while the angle-versus-length result depends on task and stimulus.

Why it happens

Aligned lengths can use common-start and endpoint cues; sectors can use boundary orientation, a full-circle reference, arc length, area, and familiar fractions. Which cue set works better changes with angle size, orientation, neighboring marks, inner and outer radius, separation, and task. Describing sector reading as a serial process of extracting boundaries and then estimating an opening is a testable mechanism hypothesis, not a substitute for behavioral results. A classic ranking is likewise not an unconditional channel property.

Studying it

Within one study, hold data, value distribution, response format, and display conditions constant while comparing sector angles, isolated segment lengths, and common-baseline bars. Separate coarse classification, ordering, ratio estimation, and exact lookup. Control chart size, stroke width, category count, labels, device, and practice while randomizing values and chart order. Report theory, original experiments, and matched replications separately, along with condition-specific error distributions and individual differences, instead of turning one mean into a fixed advantage.

Where it stops holding

Angle may be sufficient when the task is to identify the largest part, recognize rough quarter or half references, or show a few parts within a whole. Direction classification, rotation controls, and geometric angle tasks require angle because orientation is the object of interest. Direct labels can transfer exact lookup away from the angle, but they do not show that the angle itself became equally precise.

Applying it

• For exact ordering, small differences, or ratio comparison, prefer common-baseline bars or points on a common scale. Reserve angle for coarse part-to-whole or direction tasks.
• When angle is necessary, limit category count, enlarge marks and separation, label values and units directly, and review at small-screen, projection, and print sizes.
• Interaction may highlight matching items or disclose exact values, but hover cannot be the only label; keyboard and touch users need the same information.
• Practice can teach the encoding rule but should not rescue a channel mismatched to the core task. Supply a value table or structured summary as a nonvisual equivalent.

Related

• Same group: U1.03.1 Length judgment requires a common starting point · U1.03.3 Angles near horizontal or vertical are judged more accurately than oblique ones · U1.03.4 Length can only encode non-negative quantities; negatives need position · U1.03.5 Arc lengths are harder to compare than straight lengths
• Adjacent: U1.01.2 The ranking comes from magnitude-judgment experiments, not design intuition · U2.07.1 Angles read imprecisely
• Search terms: angle versus length · magnitude judgment · graphical perception