Quiz: Visualization Workflow
Part I: Multiple Choice (10 questions)
Q1. What is the chapter's threshold concept?
A) Tools over process B) Process over product C) Data over design D) Code over sketch
Answer
**B.** The quality of a visualization is determined by the process that produced it. A disciplined process with modest tools beats great tools without a process.Q2. What is the first step of the 8-step workflow?
A) Data assessment B) Chart selection C) Question D) Sketch
Answer
**C.** Question. Every visualization starts with a specific, testable question. Without it, the rest of the workflow drifts.Q3. Why is paper sketching recommended before coding?
A) Paper is cheaper than computers B) Sketching forces design decisions that coding defers to defaults C) Pen and paper produce better charts D) Matplotlib is too slow
Answer
**B.** Sketching on paper requires deciding every element (axes, labels, title, colors). Coding from scratch inherits matplotlib defaults and produces generic-looking charts.Q4. What is the purpose of the prototype step?
A) To produce a final chart B) To produce a rough version quickly for critique C) To test matplotlib features D) To impress stakeholders
Answer
**B.** The prototype is a throwaway. Its purpose is to have something concrete to critique. Over-polishing the prototype wastes time if the critique reveals the approach is wrong.Q5. Why should you use a critique rubric instead of gut feeling?
A) Rubrics are more authoritative B) Rubrics force you to check specific items you might otherwise miss C) Gut feeling is never reliable D) Rubrics are faster
Answer
**B.** A rubric structures the evaluation so specific aspects (accessibility, data integrity, etc.) are explicitly checked. Gut feeling misses systematic issues because we are blind to our own assumptions.Q6. What does the "publish" step include beyond saving the file?
A) Only exporting at the right resolution B) Format selection, branding, checklist review, distribution, and archiving C) Only applying the brand D) Only writing the caption
Answer
**B.** Publishing is a complete handoff process: format, DPI, brand, checklist, caption, distribution, and archiving the source for reproducibility.Q7. What is a common failure mode in Step 1 (Question)?
A) Too specific B) Vague or multiple conflated questions C) Too many data points D) Wrong library choice
Answer
**B.** Vague questions and multiple questions conflated into one chart are the main pitfalls. A specific, testable question is the foundation of a useful chart.Q8. When should you stop refining a chart?
A) When it is perfect B) When the question is answered clearly and no rubric item is flagged C) When the deadline hits D) When you run out of ideas
Answer
**B.** Perfectionism prevents publication. Aim for "good enough to serve the purpose." If the chart answers the question clearly and passes the rubric, ship it.Q9. What is in a personal visualization toolkit?
A) Only a matplotlib cheat sheet B) Libraries, style sheets, templates, helper modules, checklists, reference materials C) Just the Plotly documentation D) Only the workflow document
Answer
**B.** A complete toolkit includes go-to libraries, a style sheet and template, helper functions, critique rubrics, and curated reference materials (documentation, books, example galleries).Q10. Which step catches the most errors if done well?
A) Question B) Prototype C) Critique D) Publish
Answer
**C.** Critique is where systematic errors are caught. The question shapes the approach, the critique catches mistakes in execution. Most published charts have skipped or rushed the critique step.Part II: Short Answer (10 questions)
Q11. Write a specific, testable question for a dataset of daily temperatures.
Answer
"How has the annual average temperature anomaly changed between 1950 and 2024, and in which decade did the rate of change increase most sharply?" Specific (about anomalies, a specific period), testable (you can tell from the chart if it's answered), audience-neutral enough to guide chart design.Q12. Describe Step 2 (data assessment) and list the key checks.
Answer
Step 2 is understanding the data before visualizing. Checks: columns and types, row count, missing values, value ranges and distributions, temporal coverage for time series, multiple tables needing joins, data provenance. Use `df.info()`, `df.describe()`, `df.isnull().sum()`, `df.head()`.Q13. Give an example of matching a question type to a chart type.
Answer
"How has X changed over time?" → line chart. "Which category has the most Y?" → bar chart. "Do X and Y correlate?" → scatter plot with optional regression overlay. "What is the distribution of Z?" → histogram or violin plot. "Where does W occur?" → map (choropleth or dot).Q14. Write three items from the minimal critique rubric.
Answer
Any three of: (1) Question answered? (2) Data integrity? (3) Chart type appropriate? (4) Encoding appropriate? (5) Readable? (6) Accessible? (7) Title helpful? (8) Source disclosed? (9) Annotations helpful? (10) Clutter removed?Q15. Describe the difference between self-critique and group critique.
Answer
Self-critique is applying the rubric to your own work. It is valuable but limited because you are attached to your own choices. Group critique brings external perspectives that catch issues you missed. Both are valuable; group critique is more rigorous but takes more time. A mature practice uses self-critique for quick iterations and group critique for important publications.Q16. What are the main failure modes in Step 5 (Prototype)?
Answer
Over-polishing before evaluating (wastes time on the wrong direction), starting from scratch instead of adapting known patterns, premature optimization, and not running the code to verify it works. Speed is the goal at this stage; polish comes later.Q17. Describe how the workflow adapts to low-stakes vs. high-stakes charts.
Answer
For low-stakes (quick internal chart): each step compresses to seconds or minutes. Question in one sentence, data check via `df.info()`, chart type intuitive, mental sketch, quick prototype, brief critique, minimal refinement, casual publication. For high-stakes (journal figure): each step gets full attention. Extensive framing, rigorous validation, multiple sketch alternatives, careful prototype, detailed group critique, iterative refinement, polished publication with a checklist. The steps are the same; only the pace and rigor differ.Q18. Why is it valuable to keep a personal reflection on which workflow steps you tend to skip?
Answer
Awareness of your own weaknesses is the first step to fixing them. If you know you tend to skip sketching, you can make a rule to always sketch before coding. If you know you skip critique, you can schedule explicit critique sessions. The reflection turns unconscious habits into conscious choices that you can change.Q19. What does the chapter mean by "the workflow is the skill"?
Answer
Technical skills (knowing matplotlib, knowing Plotly) are necessary but not sufficient. The skill that distinguishes good practitioners from average ones is the discipline to apply a consistent process — question, sketch, critique, refine, publish. A practitioner with average tool skills but a strong process outperforms one with great tool skills but no process. The tools are interchangeable; the workflow is not.Q20. Explain the relationship between this chapter and the rest of the book.
Answer
This chapter integrates everything. Each earlier chapter taught a specific technique (perception, color, chart type, specific library, etc.). This chapter shows how those techniques fit into a process that produces actual charts. The earlier chapters are the bricks; this chapter is the mortar. Without the integration, techniques are isolated skills; with it, they become a coherent practice.Scoring Rubric
| Score | Level | Meaning |
|---|---|---|
| 18–20 | Mastery | You can apply the full workflow consistently and have built a personal toolkit. |
| 14–17 | Proficient | You know the steps; build the habit of applying them. |
| 10–13 | Developing | You grasp the process; commit to following it on your next five charts. |
| < 10 | Review | Re-read the full chapter and work through Part B exercises. |
Chapter 34 is the capstone project applying the workflow to a complete data story.