THINK FIRST·CODE LATER

← Interactive Information Visualization
Chapter 5 · Week 5

Marks, Channels and Perception

Answered 0/30 Correct 0
Sign in to save progress across devices
Q1

In a scatter plot of temperature (x) vs. PM2.5 (y) with points coloured by district, what are the marks?

Q2

Which channel is an identity channel?

Q3

Which principle is violated when district names are encoded with the size of points?

Q4

What does the effectiveness principle state?

Q5

Which magnitude channel is ranked most effective?

Q6

Why does a bar chart usually allow more accurate comparisons than a pie chart?

Q7

Cleveland and McGill (1984) studied:

Q8

According to Stevens' power law with exponent 0.7 for area, a circle with twice the area of another is perceived as about:

Q9

With Stevens' exponent 0.5 for brightness, a patch with four times the luminance looks about:

Q10

Which channel's differences are overestimated according to Stevens' exponents?

Q11

CityPulse has 12 districts. Why is colour hue a poor channel for them in one chart?

Q12

Width and height of a rectangle are perceived together as its area. In Munzner's terms, these channels are:

Q13

Which pair of channels is separable (each can be read independently)?

Q14

A single red point among 500 blue points is found almost instantly. A red square among red circles and blue squares is not. Why?

Q15

Points enclosed by a shaded region are seen as a group. Which Gestalt principle is this?

Q16

Why are comparisons easier with bars aligned on a common baseline than with floating (unaligned) segments of a stacked bar?

Q17

A bar chart of mean PM2.5 has its y axis starting at 40. The bars are 42 and 44. What is Tufte's lie factor?

Q18

When is a non-zero baseline acceptable?

Q19

Which encoding of an ordinal air-quality level (good < moderate < unhealthy) is expressive?

Q20

A treemap shows the hierarchy district → building with nested rectangles. Which way of showing relations does it use?

Q21

An AI tool produces a bubble chart in which bubble radius (not area) is proportional to trips. A station with twice the trips gets twice the radius. What is the effect?

Q22

Which chart best supports the task "compare mean PM2.5 across 6 districts precisely"?

Q23

Heer and Bostock (2010) replicated Cleveland and McGill's experiments using:

Q24

Who first formulated the expressiveness and effectiveness criteria for automatically designing graphical presentations (the APT system)?

Q25

In Vega-Lite (and Altair), the size channel of a point mark scales:

Q26

For a categorical attribute, which identity channel is ranked highest?

Q27

A heat map of 6 districts × 30 days uses a sequential colour scale for PM2.5. Which task does it support worst?

Q28 Short answer

Describe three different encodings of the same CityPulse table (date, district, PM2.5 for 6 districts over 30 days), naming marks and channels for each, the task each supports best, and one weakness of each.

Q29 Short answer

A chart uses bars with the y axis starting at 90 to show "visits" of 100 (2025) and 95 (2026). Compute the data change, the drawn change and the lie factor, and explain how you would redesign the chart honestly while still making the 5 % drop visible.

Q30 Short answer

An AI assistant proposed this encoding for CityPulse stations: x = longitude, y = latitude, point size = district (6 districts), point shape = number of trips, colour = a rainbow scale of station name (80 stations). Identify every expressiveness and effectiveness problem and propose a corrected encoding for the task "locate the busiest stations in each district".