MS-DS Master of Data science Master of Data science Data Visualization and Communication 1 — Questions and Answers
Question 1: Which color scale is most appropriate for representing data that has a meaningful midpoint, such as temperature anomalies above and below average?
- Sequential color scale
- Diverging color scale (Correct answer)
- Qualitative color scale
- Monochromatic color scale
Correct answer: Diverging color scale
Diverging color scales use two contrasting hues radiating outward from a neutral midpoint, making them ideal for data with a meaningful center value where both directions of deviation need to be visually distinguished.
Question 2: A data scientist wants to show how a company's market share changed across five product categories over three years. Which visualization is most suitable?
- Pie chart
- Scatter plot
- Grouped bar chart (Correct answer)
- Box plot
Correct answer: Grouped bar chart
A grouped bar chart allows direct comparison of multiple categories across multiple time periods simultaneously, making it easy to see both within-year category differences and year-over-year trends for each category.
Question 3: What is the primary purpose of a tooltip in an interactive data visualization?
- To filter the dataset displayed
- To reveal precise values for specific data points on hover (Correct answer)
- To highlight outliers automatically
- To zoom into a selected region of the chart
Correct answer: To reveal precise values for specific data points on hover
Tooltips provide exact underlying values when a user hovers over a data point, balancing clean overall chart design with the ability to access precise details on demand without cluttering the visual.
Question 4: When communicating uncertainty in a model's predictions to a business audience, which visual element is most commonly used?
- A rug plot along the x-axis
- Confidence interval bands or error bars (Correct answer)
- A heat map overlay
- A step chart
Correct answer: Confidence interval bands or error bars
Confidence interval bands and error bars directly encode the range of uncertainty around a central estimate, giving non-technical audiences an intuitive sense of prediction reliability without requiring statistical expertise.
Question 5: Which principle of Gestalt psychology explains why data points placed close together are perceived as belonging to the same group in a scatter plot?
- Similarity
- Closure
- Proximity (Correct answer)
- Continuity
Correct answer: Proximity
The Gestalt principle of proximity states that objects near each other are perceived as a group; in scatter plots, clustered points naturally appear as distinct groupings because of their spatial closeness.
Question 6: A presenter uses a y-axis that starts at 950 instead of 0 on a bar chart showing values between 955 and 975. What is the main problem with this choice?
- It reduces the data-ink ratio unnecessarily
- It visually exaggerates differences between bars, potentially misleading the audience (Correct answer)
- It makes the chart harder to read on mobile devices
- It prevents the use of color encoding on the bars
Correct answer: It visually exaggerates differences between bars, potentially misleading the audience
Truncating the y-axis on a bar chart makes small absolute differences appear dramatically large relative to bar heights, distorting the viewer's perception of magnitude and potentially conveying a misleading sense of significant variation.
MS-DS Master of Data science Master of Data science Data Visualization and Communication 1
Which color scale is most appropriate for representing data that has a meaningful midpoint, such as temperature anomalies above and below average?