Data Science Data Science Data Visualization and Communication Questions and Answers 2 — Questions and Answers
Question 1: Which type of chart is most appropriate for displaying the distribution of a single continuous variable?
- Histogram (Correct answer)
- Pie chart
- Scatter plot
- Treemap
Correct answer: Histogram
A histogram bins continuous values into intervals and displays their frequency, making it ideal for showing distributions.
Question 2: What does the term 'data-ink ratio' refer to in data visualization?
- The proportion of ink used to display actual data versus non-data elements (Correct answer)
- The resolution quality of a printed chart
- The ratio of colored to black-and-white elements
- The number of data points per square inch of a graphic
Correct answer: The proportion of ink used to display actual data versus non-data elements
Coined by Edward Tufte, the data-ink ratio measures how much of a graphic's ink is devoted to presenting data rather than decoration.
Question 3: When presenting model performance results to a non-technical business audience, which approach is generally most effective?
- Translate metrics into business impact such as revenue or cost savings (Correct answer)
- Show ROC curves and confusion matrices with detailed annotations
- Present raw accuracy percentages across all model iterations
- Display the full mathematical derivation of the evaluation metric
Correct answer: Translate metrics into business impact such as revenue or cost savings
Non-technical stakeholders understand business outcomes better than statistical metrics, so framing results in terms of impact is most effective.
Question 4: In a scatter plot, what does a strong negative correlation between two variables look like?
- Points trend downward from left to right (Correct answer)
- Points trend upward from left to right
- Points form a horizontal band
- Points cluster in the center of the plot
Correct answer: Points trend downward from left to right
A negative correlation means as one variable increases the other decreases, producing a downward-sloping pattern of points.
Question 5: What is the primary purpose of using a logarithmic scale on a chart axis?
- To display data that spans several orders of magnitude more clearly (Correct answer)
- To make all data points equidistant from each other
- To convert categorical data into continuous data
- To increase the number of gridlines on the axis
Correct answer: To display data that spans several orders of magnitude more clearly
A logarithmic scale compresses large ranges so that data spanning multiple orders of magnitude can be visualized without losing detail in smaller values.
Question 6: Which visualization technique is best suited for showing how a total value is broken down into parts over time?
- Stacked area chart (Correct answer)
- Box plot
- Scatter matrix
- Violin plot
Correct answer: Stacked area chart
A stacked area chart displays both the total and its component parts across a time axis, making compositional changes over time easy to see.
Data Science Data Science Data Visualization and Communication Questions and Answers 2
Which type of chart is most appropriate for displaying the distribution of a single continuous variable?