7.13: Detailed Figure Captions - Chapter 7

Data Visualization Principles -- Figure Specifications

Overview

• Chapter: 7 - Data Visualization Principles
• Total Figures Proposed: 8
• Figure Types Used: Conceptual Illustration, Comparison Table, Flowchart, Data Chart, Infographic, Architecture Diagram

---

Figure 7.1.1

Title: Preattentive Visual Attributes 

Detailed Caption: A 2×2 grid of panels — each enclosed in a thin light-gray border — fills the figure under the header "Preattentive Visual Attributes" in bold navy. The top-left panel, labeled COLOR, shows a 4×4 array of light-gray circles with a single warm-gold circle that leaps out instantly; the top-right panel, labeled SIZE, shows the same array in navy with one circle conspicuously enlarged; the bottom-left panel, labeled SHAPE, replaces one navy circle with a navy square; and the bottom-right panel, labeled ORIENTATION, tilts one vertical navy line to 45 degrees while the rest stand upright. A small teal caption beneath each panel names the detected difference — different hue, different size, different shape, different angle. By demonstrating all four attribute families on a single canvas, the figure makes the chapter's central claim tangible: the human visual system resolves color, form, orientation, and position differences in under 500 milliseconds, before conscious attention has even been deployed, and designing charts that load critical information onto these channels means the viewer's eye arrives at the insight without effort. 

Figure 7.2.1

Title: Exploratory vs. Explanatory Visualization 

Detailed Caption: A two-column matrix under the title "Exploratory vs. Explanatory Visualization" organizes five rows of comparison beneath color-coded header pills — a navy "Exploratory Visualization" pill on the left (with a magnifying-glass icon) and a teal "Explanatory Visualization" pill on the right (with a presentation icon). Row labels run down the left margin in bold: Goal, Audience, Design, Data Scope, and Typical Output. Reading across, the contrast sharpens with each row: the exploratory column pairs "Discover patterns & insights" with a rough, iterative, many-charts workflow aimed at the analyst alone, while the explanatory column pairs "Communicate specific findings" with a polished, annotated, stakeholder-facing output. A warm-gold accent line divides the two columns, and alternating light-teal and white row backgrounds make horizontal scanning effortless. The "Data Scope" row crystallizes the chapter's core metaphor — "exploring 100 rocks" on the left versus "presenting 1–2 gemstones" on the right — making vivid why the same dataset demands completely different design decisions depending on whether the analyst is still searching or has already found what they need to say. 

Figure 7.3.1

Title: Chart Type Selection Guide 

Detailed Caption: A top-to-bottom decision flowchart opens at a single navy rounded rectangle asking "What do you want to show?" and immediately fans into five teal branches, each naming a distinct analytical intent: Comparison across categories, Trend over time, Relationship between variables, Part of a whole, and Patterns in a matrix. Each branch terminates in a warm-gold recommendation tile — Bar Chart, Line Chart, Scatter Plot, Pie Chart (use sparingly), and Heatmap respectively — each accompanied by a miniature icon and a short gray tip printed below: "Start y-axis at 0," "Show direction of change," "Add trend line if needed," "Max 5–6 slices," and "Order rows/columns logically." Connector arrows are clean and rounded; yes/no logic is replaced by intent labels so the question the analyst asks in their head maps directly onto the branch they follow. By converting chart selection from a memorization task into a single-question decision tree, the figure gives students a portable lookup they can apply mid-analysis — identify what the chart must communicate, follow the matching branch, and the appropriate chart type (with its most important constraint) sits at the end of the path. 

Figure 7.4.1

Title: Dimensions vs. Measures in Data Visualization

Detailed Caption: A three-part horizontal layout moves left to right across the figure, connected by navy arrows. On the left, a six-row sample data table displays three columns: Region and Quarter, whose headers are highlighted in teal under the label "DIMENSIONS (categorical, GROUP BY)," and Sales ($), whose header is highlighted in warm gold under the label "MEASURE (numeric, SUM)." A center arrow labeled "Aggregate & Plot" in navy carries the eye to the right panel, where a simple bar chart shows three navy bars — North, South, and East — with the x-axis labeled "Dimension" in teal and the y-axis labeled "Sum of Sales" in warm gold. A summary line printed beneath the entire figure reads "Dimensions slice the data; Measures quantify it." The spatial separation between the raw table and the finished chart, bridged by a single labeled aggregation step, makes the otherwise abstract mapping from data structure to chart geometry concrete: students can trace any column in the table to its role in the chart and understand that the distinction between a dimension and a measure is not a software concept but a decision about whether a field names a group or quantifies one. 

Figure 7.5.1

Title: Data Storytelling Framework 

Detailed Caption: Three large chevron-connected stages span the figure horizontally under the banner "Data Storytelling: Beginning → Middle → End" in navy. The first stage, on a teal background, is labeled "1. CONTEXT (Setup)" and bullets three preparatory actions — Define the question, Introduce the data source, Set audience expectations — beneath a small document-with-question-mark icon. The second stage, on a navy background, is labeled "2. EVIDENCE (Visual Insight)" and bullets three presentation actions — Present the key chart, Use preattentive attributes to highlight findings, Annotate critical data points — beneath a bar-chart-with-one-highlighted-bar icon. The third stage, on a warm-gold background, is labeled "3. INTERPRETATION (Action)" and bullets three closing actions — State what the data means, Provide recommendations, Connect to audience needs — beneath a lightbulb icon. A gray italic quote anchors the bottom of the figure: "63% of people remember stories vs. 5% who remember statistics alone." By mapping the classical three-act narrative arc onto a data presentation workflow, the figure reframes every chart as not an endpoint but a middle act — useful only when the audience has been oriented by context first and guided toward action by interpretation afterward. 

---

Figure 7.6.1

Title: Design Principles for Effective Visualization 

Detailed Caption: A 2×2 grid of rounded-corner cards fills the figure, each card dedicated to one of four foundational design principles. The top-left card carries a teal header bar with an eye icon and the label "CLARITY," listing four guidelines in dark gray body text: Remove chartjunk, Label axes with units, Choose the right chart type, and Make insight obvious at a glance. The top-right card carries a navy header bar with a chain-link icon and the label "CONSISTENCY," listing four guidelines: Same color = same category across charts, Uniform axis scales for comparison, Consistent fonts and formatting, and Reduces cognitive load. The bottom-left card carries a warm-gold header bar with a universal-access icon and the label "ACCESSIBILITY," listing four guidelines including the reminder that 8% of men have color vision deficiency and that patterns plus color — not color alone — should encode categories. The bottom-right card carries a teal-to-navy gradient header with a palette icon and the label "COLOR USAGE," listing four guidelines for purposeful palette selection: limit to five or six distinct colors, and choose sequential palettes for ordered data versus diverging palettes for positive-and-negative deviations. By consolidating all four principles into a single scannable grid, the figure functions as a design checklist — students can hold it alongside any chart they build and ask, card by card, whether each principle has been satisfied. 

Figure 7.7.1

Title: Visualization Tools Landscape

Detailed Caption: A four-column comparison matrix organizes key tool attributes beneath navy header pills for Matplotlib, Seaborn, Plotly, and Tableau, each pill carrying a small logo-style icon. Six row labels in bold teal run down the left margin — Type, Language, Interactivity, Learning Curve, Best For, and Customization — and each cell carries a concise phrase that characterizes the tool on that dimension. Warm-gold filled circles serve as a visual rating system under Learning Curve (one circle for easiest, three for steepest), providing an at-a-glance difficulty signal alongside the textual descriptor. Alternating white and very light gray row backgrounds assist horizontal scanning across four columns. The "Best For" row is the most differentiated: Matplotlib for custom publication-quality charts, Seaborn for quick statistical plots, Plotly for web dashboards and interactive exploration, and Tableau for business dashboards and reporting. By mapping all four tools across the same six dimensions simultaneously, the figure saves students from the common mistake of defaulting to whichever tool they learned first — the matrix makes the trade-offs between customization, interactivity, language requirement, and ease of use legible in a single reading. 

 

Figure 7.8.1

Title: Common Visualization Mistakes 

Detailed Caption: Three horizontal rows, each pairing a red-X "WRONG" example on the left with a green-checkmark "RIGHT" corrected version on the right separated by a directional arrow, illustrate the three most consequential visualization mistakes. Row 1, labeled "Truncated Y-Axis," shows a bar chart of two bars — values 48 and 52 — where the y-axis begins at 45, making a 4-unit difference appear enormous, corrected on the right by a y-axis starting at 0 that reveals the difference is negligible. Row 2, labeled "Overloaded Chart," shows a pie chart with twelve or more tiny, overlapping, illegible slices on the left, corrected on the right by a clean horizontal bar chart showing the top five categories plus an "Other" grouping. Row 3, labeled "3D Distortion," shows a 3-D bar chart where perspective foreshortening makes rear bars appear shorter than their actual values, corrected on the right by a flat 2-D bar chart where every bar's height is proportional and accurate. Navy is used for correct chart elements, warm gold for highlighted correct features, and red and green exclusively for the X and checkmark verdict markers — keeping the color vocabulary purposeful rather than decorative and ensuring the figure itself models the design principles it is teaching.