Seeing Data: Visualization for Impact
Data Visualization Masterclass
Master the art and science of data visualization for development. From perceptual foundations to interactive dashboards—learn to communicate data that drives decisions and creates change.
Interactive Tools
AI Study Companion
Colab Notebooks
58-Term Lexicon
Course Papers Collection
Why Visualize? Purpose & Principles
Understanding when and why visualization matters for policy influence and accountability. From Anscombe's quartet to Hans Rosling's Gapminder—discover the power of seeing data.
The Case for Visualization
Data visualization is not decoration—it is a cognitive tool that extends human perception. Our visual system processes images in parallel, detecting patterns, outliers, and relationships that would take hours to discover in tables of numbers.
Anscombe's Quartet: The Classic Demonstration
In 1973, statistician Francis Anscombe created four datasets with nearly identical statistical properties (same mean, variance, correlation, and regression line)—yet radically different when visualized. This quartet remains the definitive argument for always visualizing your data before analysis.
Development Sector Implication
Program data often hides critical patterns. A coverage rate of 75% looks identical whether it's evenly distributed or concentrated in accessible areas. Visualization reveals the difference between success and inequity.
Foundational Frameworks
This course draws on four major intellectual traditions in data visualization, each offering distinct principles that together form a complete approach.
Edward Tufte
Data-ink ratio, chartjunk elimination, small multiples, sparklines. "Above all else, show the data."
Alberto Cairo
The Five Qualities: Truthful, Functional, Beautiful, Insightful, Enlightening—in that order of priority.
Cole Nussbaumer Knaflic
Storytelling with data: context, clutter elimination, attention direction, and narrative structure.
Data Feminism
D'Ignazio & Klein: Examine power, challenge binaries, elevate emotion, make labor visible, embrace pluralism.
Tufte's Core Principles
Definition: The proportion of a graphic's ink devoted to the non-redundant display of data-information. Tufte advocates maximizing this ratio by erasing non-data-ink and redundant data-ink.
| Principle | Description | Application |
|---|---|---|
| Data-Ink Ratio | Maximize ink used for data; minimize everything else | Remove gridlines, borders, backgrounds that don't aid comprehension |
| Chartjunk | Decorative elements that don't convey information | Eliminate 3D effects, clip art, unnecessary icons, heavy gridlines |
| Lie Factor | Size of effect in graphic ÷ size of effect in data | Ensure visual proportions match numerical proportions exactly |
| Small Multiples | Repeat same chart structure across categories/time | Compare districts, years, or groups using consistent layouts |
| Sparklines | Intense, word-sized graphics embedded in text | Show trends inline without breaking reading flow |
Cairo's Five Qualities
Alberto Cairo's hierarchy establishes that visualization must be truthful first. Beauty without truth is deception; function without insight is mere display.
Essential Reading
Exercise 1.1: Data-Ink Audit
Find three visualizations from recent UNDP, World Bank, or NGO annual reports. For each, identify: (1) elements that could be removed without losing meaning, (2) approximate data-ink ratio, and (3) one improvement that would increase clarity while reducing visual noise.
Book 1:1 Coaching
Essential Reading
Related ImpactMojo Resources
Data Feminism 101
Check Your Understanding
Test your comprehension of the key concepts from this module.
Think of a chart, map, or visualization that genuinely changed how you understood a topic. What made it effective? What design choices enabled that insight?
Data Types & Quality Structures & Challenges
Classify data correctly, recognize development sector data quality challenges, and understand how data collection choices constrain visualization options.
Data Type Classification
Every visualization decision begins with understanding your data type. The encoding that works for quantitative data fails for categorical; the chart perfect for time series misleads for cross-sectional comparisons.
| Data Type | Definition | Examples | Suitable Encodings |
|---|---|---|---|
| Nominal | Categories without order | Country names, program types, gender categories | Position, color hue, shape |
| Ordinal | Categories with meaningful order | Education levels, satisfaction ratings, wealth quintiles | Position, color saturation, size |
| Interval | Numeric with arbitrary zero | Temperature (°C), dates, index scores | Position, length, area (with caution) |
| Ratio | Numeric with true zero | Income, population, coverage rates | Position, length, area, angle |
| Temporal | Time-based sequences | Years, months, project phases | Position on x-axis, animation |
| Geographic | Spatial coordinates or regions | Districts, GPS points, administrative boundaries | Map position, choropleth, symbols |
Development Sector Data Challenges
Development data presents challenges rarely addressed in standard visualization courses. Acknowledging these constraints is essential for honest, effective communication.
Missing Values
Subnational data gaps, incomplete time series, non-response. Never interpolate without disclosure.
Methodology Changes
Redefined indicators, updated survey instruments, changed sampling. Breaks in time series must be marked.
Aggregation Masking
National averages hide regional inequality. Always ask: "Does this aggregate conceal important variation?"
Proxy Indicators
What we can measure ≠ what we want to measure. Document the gap between proxy and concept.
Small Samples
Disaggregated data often lacks statistical power. Confidence intervals are not optional.
Reporting Delays
"Latest data" may be 2-3 years old. Always show data vintage prominently.
J-PAL Principle: Automate Everything
Manual copying between software introduces errors. Code-based pipelines (Stata → LaTeX, R Markdown, Python notebooks) ensure data provenance and reproducibility. If you can't trace a number back to its source, don't publish it.
IASC Information Management Principles
The Inter-Agency Standing Committee establishes standards for humanitarian data that apply broadly to development sector visualization.
Core Principles
- Interoperability: Data can be combined across sources
- Accountability: Clear ownership and responsibility
- Verifiability: Claims can be traced to evidence
- Relevance: Information serves decision-making needs
Visualization Implications
- Always cite data sources with retrieval dates
- Document any transformations applied
- Show uncertainty where it exists
- Design for the decisions your audience makes
Exercise 2.1: Data Quality Audit
Download a DHS (Demographic and Health Survey) dataset for any country. Document data quality issues: missing disaggregations, definitional changes over time, geographic coding inconsistencies. Propose visualization approaches that honestly represent uncertainty.
Practice Datasets
These datasets are ideal for learning visualization—well-documented, freely available, and relevant to development work.
Stats Assistant Tool
Essential Reading
Related ImpactMojo Resources
EDA for Surveys
Qualitative Methods
Check Your Understanding
Test your comprehension of the key concepts from this module.
Consider a dataset you've worked with recently. What data quality issues did you encounter (missing values, inconsistent formats, outliers)? How did you address them, and how might those choices affect visualization?
Visual Encoding Graphical Perception
The scientific foundation for chart selection. Rank visual encodings by perceptual accuracy and predict which chart types will enable accurate comparisons.
Cleveland & McGill's Hierarchy
In 1984, William Cleveland and Robert McGill conducted experiments establishing a hierarchy of encoding accuracy. Their findings explain why bar charts consistently outperform pie charts: humans judge length on a common baseline more accurately than angles.
From most to least accurate perception: Position on common scale → Position on non-aligned scales → Length → Angle/Slope → Area → Volume → Color saturation/density.
Why This Matters
A pie chart forces viewers to compare angles—ranked 4th in accuracy. A bar chart uses position on a common scale—ranked 1st. The bar chart isn't just "better design"—it's perceptually easier for human brains to decode accurately.
Mackinlay's Extensions
Jock Mackinlay extended Cleveland and McGill's work to different data types, recognizing that optimal encodings differ for quantitative, ordinal, and nominal data.
| Encoding | Quantitative | Ordinal | Nominal |
|---|---|---|---|
| Position | ★★★ Excellent | ★★★ Excellent | ★★★ Excellent |
| Length | ★★★ Excellent | ★★ Good | ★ Poor |
| Angle | ★★ Good | ★ Poor | ★ Poor |
| Area | ★★ Good | ★★ Good | ★ Poor |
| Color Saturation | ★ Poor | ★★ Good | ★ Poor |
| Color Hue | ★ Poor | ★ Poor | ★★★ Excellent |
| Shape | ★ Poor | ★ Poor | ★★ Good |
Gestalt Principles in Data Visualization
Gestalt psychology explains how humans perceive visual groupings. These principles guide layout decisions that make data relationships intuitive.
Proximity
Elements close together are perceived as related. Group related data points; separate unrelated ones.
Similarity
Elements that look alike are perceived as part of the same group. Use consistent colors for same categories.
Enclosure
Elements within boundaries are perceived as grouped. Use backgrounds or borders to organize content.
Connection
Lines between elements create perceived relationships. Stronger than proximity for showing links.
Continuity
Elements aligned are perceived as related. Align elements that should be compared.
Closure
The eye completes incomplete shapes. You can remove unnecessary chart frames—viewers will perceive them.
Exercise 3.1: Perception Experiment
Recreate a simplified Cleveland-McGill experiment with classmates. Show pairs of values encoded as: (a) bars, (b) pie slices, (c) circles by area. Compare estimation accuracy across conditions. Document your findings and explain why certain encodings performed better.
Essential Reading
Related Resources
D3 Chart Gallery
Check Your Understanding
Test your comprehension of the key concepts from this module.
Find a 3D pie chart or bubble chart in a report or news article. How could you redesign it using more effective visual encodings (position, length) to communicate the same information more accurately?
Color & Accessibility Inclusive Design
Select appropriate palettes for sequential, diverging, and categorical data. Design for colorblind accessibility and adapt for cultural contexts.
Three Types of Color Palettes
Color palette selection is not aesthetic preference—it encodes information. Using the wrong palette type misleads viewers by implying relationships that don't exist.
Sequential
Light → Dark. For ordered data with no midpoint: population density, poverty rates, coverage percentages.
Diverging
Color ← Neutral → Color. For data with meaningful midpoint: change from baseline, deviation from target, above/below average.
Categorical
Distinct hues. For unordered groups: regions, program types, demographic categories. Maximum ~7-8 distinguishable colors.
Never Use Rainbow Palettes
Rainbow color scales (common in GIS outputs) create false boundaries, lack perceptual ordering, and fail completely for colorblind viewers. There is no circumstance where rainbow is the right choice.
Colorblind-Friendly Design
Approximately 8% of males and 0.5% of females have some form of color vision deficiency. Designing for colorblindness isn't accommodation—it improves clarity for everyone.
Common Colorblindness Types
- Deuteranopia: Cannot distinguish red-green (most common)
- Protanopia: Reduced sensitivity to red
- Tritanopia: Cannot distinguish blue-yellow (rare)
- Achromatopsia: Complete color blindness (very rare)
Design Solutions
- Use Viridis palette (perceptually uniform, colorblind-safe)
- Avoid pure red-green combinations
- Add texture, patterns, or direct labels
- Test with Color Oracle or Sim Daltonism
- Ensure grayscale differentiation
The Viridis family (viridis, magma, plasma, inferno, cividis) are designed to be: perceptually uniform (equal steps in data = equal steps in perception), robust to colorblindness, and print well in grayscale. They are the default in ggplot2 and matplotlib for good reason.
Cultural Considerations
The WHO Data Design Language emphasizes that color carries cultural meaning. Development visualizations must be "truly internationalized, multilingual, culturally aware."
| Color | Western Association | Alternative Associations | Guidance |
|---|---|---|---|
| Red | Danger, stop, negative | Prosperity, celebration (China, India) | Avoid for negative meaning in Asian contexts |
| Green | Go, positive, nature | Islamic significance; sometimes death (S. America) | Consider religious sensitivities |
| White | Purity, cleanliness | Mourning, death (many Asian cultures) | Test with target audience |
| Blue | Trust, calm, corporate | Generally positive globally | Usually safe choice |
Essential Tools
Exercise 4.1: Palette Redesign
Find a rainbow-colored choropleth map from any UN agency report. Redesign it using an appropriate sequential palette from ColorBrewer. Test both versions with a colorblindness simulator and document the difference in readability.
Essential Reading
ColorBrewer
Coblis Simulator
Color Oracle (Desktop)
Check Your Understanding
Test your comprehension of the key concepts from this module.
Take a colorful visualization and test it with a colorblindness simulator. What information is lost? How would you redesign it to be accessible while maintaining visual appeal?
Chart Selection Frameworks & Common Mistakes
Use evidence-based frameworks to select chart types. Identify and correct the most common visualization errors found in development sector reports.
The Grammar of Graphics
Leland Wilkinson's Grammar of Graphics (1999) provides the theoretical foundation underlying ggplot2, Vega-Lite, and modern visualization thinking. Understanding this grammar enables creating novel chart types, not just selecting from menus.
Any visualization decomposes into: Data → Transformations → Coordinate System → Scales → Geometric Elements → Guides → Facets. Master these building blocks and you can construct any chart.
Chart Chooser Frameworks
Several excellent frameworks help match data and purpose to chart type. Each has strengths for different contexts.
| Framework | Best For | Access |
|---|---|---|
| From Data to Viz | Technical users; organized by input data type | data-to-viz.com |
| FT Visual Vocabulary | Storytelling; 9 categories with D3 templates | FT GitHub |
| Evergreen's Qualitative Chooser | Interview/focus group data visualization | stephanieevergreen.com |
| PolicyViz Graphic Continuum | 90+ chart types beyond basics | PolicyViz |
| Andrew Abela's Diagram | Beginners; simple decision tree | Search "Abela chart chooser" |
The Seven Deadly Chart Sins
These errors appear repeatedly in development sector reports. Learn to identify and avoid them.
1. Truncated Y-Axes
Starting y-axis above zero exaggerates small differences. Reserve for financial data where small changes matter greatly.
2. Pie Charts with 10+ Slices
Humans can distinguish ~7 categories max. Beyond that, pie charts become unreadable. Use bar charts instead.
3. 3D Effects
3D distorts perception and adds zero information. There is no valid use case for 3D in 2D data display.
4. Dual Y-Axes
Two scales on one chart implies correlation where none may exist. Use two separate charts with aligned x-axes.
5. Rainbow Color Scales
Creates false boundaries, lacks perceptual ordering, fails for colorblind viewers. Never appropriate.
6. Chartjunk
Decorative icons, heavy gridlines, unnecessary gradients, clip art. If it doesn't encode data, remove it.
7. Missing Context
No benchmarks, comparisons, or uncertainty. A number without context is meaningless. Always show: compared to what?
Tables vs. Charts: When to Use Each
Stephen Few's guidance: Use tables when precise lookup of individual values matters; use charts when the message is in the shape—patterns, trends, outliers.
Use Tables When...
- Readers need to look up specific values
- Precision matters (exact numbers, not patterns)
- You're comparing items across multiple attributes
- The dataset is small (under 20 rows)
- Readers will refer back repeatedly
Use Charts When...
- The message is about shape, trend, or pattern
- You want to show relationships between variables
- Comparison of magnitudes is the goal
- Dataset is large and tables would overwhelm
- Emotional impact matters for advocacy
Exercise 5.1: Chart Makeover
Collect three visualizations from recent development sector reports (NGO annual reports, World Bank publications, UN agency materials). Identify errors using the seven-mistake framework. Redesign one visualization addressing all identified issues.
Chart Selection Frameworks
Use these evidence-based decision tools instead of defaulting to whatever Excel suggests:
Request Workshop
Essential Reading
- Knaflic, C.N. (2015). Storytelling with Data, Ch. 2: Choosing an Effective Visual
- Schwabish, J. (2021). Better Data Visualizations, Part II: Chart Types
- Few, S. (2004). "Choosing a Chart Type" — Show Me the Numbers excerpt [PDF]
- Schwabish, J. — "Five Charts You've Never Used But Should" [Video, 15 min]
Related ImpactMojo Resources
Storytelling Lab
Check Your Understanding
Test your comprehension of the key concepts from this module.
For each question, identify the best chart type: (1) How has GDP changed over 50 years? (2) What share of budget goes to each department? (3) Is there a relationship between education and income? Justify your choices.
Design Process Style Guides & Iteration
Apply iterative design methodology. Develop organizational style guide components that institutionalize visualization practice beyond individual skill.
The Design Sprint Methodology
Following Harvard CS171's approach, effective visualization emerges from structured iteration: paper sketching before digital tools, multiple alternatives before committing, structured critique at each stage.
1. Sketch
Paper prototypes first. No software until you've explored 3+ approaches by hand.
2. Prototype
Build functional draft in your tool of choice. Focus on data, not polish.
3. Critique
Structured feedback using specific criteria. Not "I like/don't like" but evidence-based review.
4. Refine
Iterate based on feedback. Polish only after structure and message are validated.
Organizational Style Guides
Individual visualization skill is insufficient. Organizations need consistent standards that enable anyone to create on-brand, high-quality outputs.
"No one should be able to tell what software you used—design quality matters more than tool choice."— Stephanie Evergreen, Data Visualization Academy
Style Guide Components
| Component | What to Define | Example Reference |
|---|---|---|
| Color Palette | Primary, secondary, sequential, diverging, categorical; hex codes | Urban Institute Style Guide |
| Typography | Title, subtitle, axis labels, annotations; sizes and weights | BBC Visual Journalism |
| Chart Types | Approved types with usage guidance; when to use each | FT Visual Vocabulary |
| Logo/Branding | Placement, clear space, co-branding rules | Organization brand book |
| Accessibility | Colorblind testing, alt text, minimum contrast ratios | WCAG 2.1 guidelines |
| DEI Considerations | Inclusive imagery, avoiding stereotypes, representation | Urban Institute DEI appendix |
Model Style Guides to Study
Exercise 6.1: Mini Style Guide
Draft a 5-page visualization style guide for a hypothetical development organization, including: color palette (with accessibility testing), 3-4 recommended chart types with examples, typography specifications, and logo placement guidelines. Use an existing style guide as template.
Check Your Understanding
Test your comprehension of the key concepts from this module.
For an organization you work with (or a fictional one), define the basics of a visualization style guide: primary and secondary colors, fonts for titles and labels, and one rule for chart design. Explain each choice.
Storytelling Narrative & Audience
Apply Cole Nussbaumer Knaflic's six lessons systematically. Design for specific audiences— donors, communities, policymakers—with appropriate narrative structures.
The Six Lessons of Storytelling with Data
Cole Nussbaumer Knaflic's framework provides a systematic approach to data communication that transforms charts into stories.
1. Understand Context
Who is your audience? What do they need to know or do? What's the desired action?
2. Choose Display
Match visualization to message: simple text for 1-2 numbers, lines for trends, bars for comparisons.
3. Eliminate Clutter
Reduce cognitive load. Remove non-strategic elements that don't serve the message.
4. Draw Attention
Use preattentive attributes (color, size, position) strategically to guide the eye.
5. Think Like Designer
Affordances, accessibility, visual hierarchy. Design for how people actually perceive.
6. Tell a Story
Beginning-middle-end structure. Narrative arc with tension and resolution.
The Big Idea Exercise
Before designing, distill your message to one sentence that captures the "so what?" This forces clarity and ensures every design choice serves the core message.
If you had only 3 minutes to explain your analysis to a busy executive, what would you say? Write it down. This becomes your narrative structure. Everything in your visualization should support this story—if it doesn't, remove it.
Designing for Different Audiences
Development sector visualizations serve diverse stakeholders with different needs, expertise levels, and decision-making contexts.
Donor-Facing
Emphasize ROI, outcomes achieved, value for money. Clear metrics, benchmarks, cost-effectiveness. Professional, polished aesthetic.
Community-Facing
Center lived experience and agency. Simple, accessible, local language. Focus on what matters to community, not donor priorities.
Policy-Facing
Actionable evidence with clear implications. Show options and trade-offs. Credible sources, uncertainty acknowledged.
Levels of Data Communication
Jon Schwabish distinguishes between annotation, narration, and true storytelling— each appropriate for different contexts.
| Level | Definition | When to Use |
|---|---|---|
| Annotation | Labels, titles, axis descriptions on charts | All visualizations—minimum required for understanding |
| Narration | Text explaining what viewers see in the data | Reports, presentations, dashboards with explanation |
| True Storytelling | Emotional connection, tension, resolution—narrative arc | Advocacy, annual reports, public communication |
Exercise 7.1: Three Audiences
Take a single dataset (e.g., vaccination coverage by district). Create three different visualizations for three audiences: (1) Ministry of Health officials needing to allocate resources, (2) community health workers identifying priority households, (3) international donors assessing program effectiveness. Write one-paragraph justification for each design choice.
Essential Reading
Related ImpactMojo Resources
Advocacy Basics
Check Your Understanding
Test your comprehension of the key concepts from this module.
Take a dataset you know well. Outline a three-part data story: (1) Hook—what surprising finding will grab attention? (2) Build—what context and evidence supports it? (3) Call to action—what should the audience do?
Tool Landscape Selection & Evaluation
Survey the visualization tool ecosystem. Evaluate tools against specific use cases and understand when each is appropriate.
Tool Selection Matrix
Every tool has trade-offs. The right choice depends on your skill level, output format, reproducibility needs, and organizational context.
| Tool | Best For | Limitations | Learning Curve |
|---|---|---|---|
| Excel / Google Sheets | Quick analysis, familiar interface, print reports | Limited chart types, manual formatting | Low |
| Tableau | Interactive dashboards, rapid exploration | Expensive licensing; free tier publishes all work publicly | Medium |
| Power BI | Microsoft ecosystem, DAX calculations | Steeper learning curve, Windows-centric | Medium-High |
| Datawrapper | News-style responsive charts, beginners | Limited customization, primarily static | Low |
| Flourish | Scrollytelling, animations, bar chart races | Templates can be constraining | Low-Medium |
| RAWGraphs | Unconventional chart types, SVG export | No hosting, static only, requires design skills | Low |
| R (ggplot2) | Publication-quality graphics, reproducibility | Programming required | High |
| Python (matplotlib, seaborn, plotly) | Data science workflows, dashboards | Multiple libraries to learn | High |
| D3.js | Custom interactive web visualization | JavaScript required, steep learning curve | Very High |
The Scaffolding Approach
Stephanie Evergreen recommends a progression: start with what's familiar (Excel), build to more powerful tools (Tableau/Flourish), optionally extend to code-based approaches (R/Python) for reproducibility.
The Invisible Tool Principle
"No one should be able to tell what software you used." Design quality matters more than tool sophistication. A well-crafted Excel chart beats a poorly designed Tableau dashboard.
Development Sector Tool Considerations
Development contexts present unique constraints: offline access, limited bandwidth, multilingual needs, and organizational licensing limitations.
Constraints to Consider
- Many field offices lack reliable internet
- PDF/print outputs remain essential
- Expensive licensing may not be available
- Staff turnover requires accessible tools
- Multi-language support is often needed
Recommended Approach
- Master Excel first—it's everywhere
- Add Datawrapper/Flourish for web outputs
- Build R/Python capacity for reproducibility
- Document everything in organizational style guide
- Create templates others can adapt
Exercise 8.1: Tool Comparison
Create the same visualization (life expectancy trends for 5 countries) in three different tools: Excel, Datawrapper, and one of Tableau/Flourish/R. Compare: time required, customization flexibility, output quality, reproducibility. Document your findings.
Try These Tools: Interactive Notebooks
Open these notebooks in Google Colab to practice with real development datasets—no installation required!
Tool Resources
Essential Reading & Videos
Check Your Understanding
Test your comprehension of the key concepts from this module.
For each scenario, recommend a tool: (1) Quick executive dashboard updated weekly (2) Custom interactive visualization for a journalism piece (3) Reproducible analysis for academic publication. Justify each recommendation.
Interactive Visualization Web & Dynamic
Implement Shneiderman's visual information-seeking mantra. Create interactive web visualizations and understand when interactivity adds value versus complexity.
Shneiderman's Mantra
Ben Shneiderman's information-seeking mantra guides effective interactive design: "Overview first, zoom and filter, then details-on-demand."
1. Overview First
Start with the big picture. Let users see the entire dataset or summary before drilling down.
2. Zoom & Filter
Enable users to focus on subsets of interest. Region, time period, category—progressive refinement.
3. Details on Demand
Hover, click, or tap to reveal specifics. Don't clutter the view—reveal context when requested.
The Grammar of Graphics in Practice
Whether using ggplot2 (R), Vega-Lite (declarative JSON), or Observable Plot, understanding the grammar enables creating any visualization from first principles.
// Vega-Lite specification example
{
"data": { "url": "health_data.csv" },
"mark": "bar",
"encoding": {
"x": { "field": "district", "type": "nominal" },
"y": { "field": "coverage", "type": "quantitative" },
"color": { "field": "region", "type": "nominal" }
}
}
Grammar Components
| Component | Purpose | Examples |
|---|---|---|
| Data | The dataset to visualize | CSV, JSON, database query |
| Marks | Geometric elements representing data | Points, bars, lines, areas |
| Encodings | Mappings from data to visual properties | x, y, color, size, shape |
| Scales | Transformations from data to visual domain | Linear, log, ordinal, time |
| Guides | Visual aids for interpretation | Axes, legends, annotations |
| Facets | Small multiples by category | Grid by region, row by year |
When to Add Interactivity
Interactivity is not always beneficial. It adds complexity, requires maintenance, and may exclude users with accessibility needs or slow connections.
Add Interactivity When...
- Dataset is too large for static display
- Users have different questions to explore
- Details need to be available without cluttering
- Comparison across multiple dimensions is needed
- Users will return repeatedly to monitor changes
Keep it Static When...
- Message is simple and single-purpose
- PDF/print output is primary use case
- Audience has limited technical access
- Interactivity would distract from key message
- Resources for maintenance are limited
Exercise 9.1: Interactive Prototype
Build an interactive visualization of SDG indicator data using Vega-Lite, Flourish, or Observable Plot that implements: (1) filtering by region, (2) hover details, (3) linked highlighting across charts. Publish to Observable or embed in a web page.
Essential Reading & Videos
Check Your Understanding
Test your comprehension of the key concepts from this module.
Sketch an interactive dashboard for monitoring a program or project. What filters would users need? What drill-down capabilities? How would views be linked? Consider both expert and casual users.
M&E Dashboards Development Applications
Design dashboards following Stephen Few's principles. Apply DHIS2 and WHO standards for M&E frameworks, theory of change communication, and annual reporting.
Dashboard Definition
Stephen Few defines a dashboard as: "A visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so information can be monitored at a glance."
Key Insight
A dashboard is not a report with charts. It's a monitoring interface for ongoing decision-making. If users need to scroll or click through pages, you've built a report, not a dashboard.
Dashboard Levels in Development
Development sector dashboards serve multiple levels of decision-making. Each requires different metrics, update frequency, and design approach.
Strategic Level
3-5 impact KPIs, outcome trends, geographic coverage. For leadership and board.
Tactical Level
Output monitoring, activity progress, data quality metrics. For M&E officers and managers.
Operational Level
Facility/site data, beneficiary tracking, financial status. For field coordinators.
DHIS2 Dashboard Standards
DHIS2 is used by 100+ ministries of health worldwide. Its dashboard standards represent field-tested best practices for health information systems.
| Principle | Implementation | Rationale |
|---|---|---|
| Indicator-Based | Show calculated indicators, not raw data | Users need actionable metrics, not data dumps |
| Relative Periods | "Last 12 months" not "Jan-Dec 2024" | Dashboards remain relevant without manual updates |
| Comparison Built-In | Show trends, targets, and benchmarks | Numbers without context are meaningless |
| Drill-Down Enabled | Click to see subnational detail | Supports Shneiderman's mantra |
Theory of Change Visualization
Theory of change diagrams present unique challenges. UNICEF research notes: "It's too easy to create visual diagrams that fail to provide a coherent picture."
Common ToC Mistakes
- Arrows everywhere without clear logic
- Boxes of inconsistent granularity
- Assumptions hidden or missing
- No timeline or sequencing
- Diagram contradicts narrative
Effective ToC Design
- Show causal links explicitly
- Document assumptions visually
- Include realistic timelines
- Pair diagram with narrative explanation
- Test with stakeholders for clarity
Exercise 10.1: Dashboard Mockup
Design a dashboard mockup for a hypothetical health program with three user personas: (1) Country Director needing monthly progress overview, (2) M&E Officer needing data quality monitoring, (3) Field Coordinator needing facility-level detail. Use Figma, PowerPoint, or paper prototypes.
Related ImpactMojo Resources
Logframe Lab
Essential Reading
Check Your Understanding
Test your comprehension of the key concepts from this module.
Find an M&E or development dashboard online (World Bank, UNDP, government). What does it do well? What could be improved? How well does it serve different audiences (program managers vs. policymakers vs. public)?
Advanced Topics Specialized Approaches
Create effective geographic visualizations, network diagrams, and apply Data Feminism principles to challenge power dynamics in visualization.
Geographic Visualization
Maps are powerful but dangerous. Choropleth maps (color-filled regions) create visual bias toward large regions regardless of population.
Choropleth Maps
Best for: rates, percentages, density. Caution: large areas dominate visually regardless of importance.
Proportional Symbols
Best for: counts, totals. Circles sized by value placed at geographic points.
Cartograms
Distort geography to size by variable (population, GDP). Corrects area bias but sacrifices familiarity.
Hex Bin Maps
Equal-area hexagons for fair comparison. Used for electoral maps (each hex = one district).
Data Feminism Principles
Catherine D'Ignazio and Lauren Klein challenge standard visualization approaches to center equity, power analysis, and diverse perspectives.
| Principle | Application to Visualization |
|---|---|
| Examine Power | Whose perspective frames the data? Who benefits from this representation? |
| Challenge Binaries | Question male/female and other categorical assumptions in data collection. |
| Elevate Emotion | Affective visualization can be rigorous. Feeling is a form of knowing. |
| Make Labor Visible | Credit data collectors, acknowledge communities who provided information. |
| Embrace Pluralism | Multiple valid interpretations exist. Avoid claiming objective truth. |
| Consider Context | Data are never neutral. Show the conditions under which data were produced. |
"Data are always collected by humans, processed by humans, and interpreted by humans. The goal is not to eliminate this human element but to make it visible."— Catherine D'Ignazio & Lauren Klein, Data Feminism
Case Study: Warming Stripes
Ed Hawkins' Warming Stripes exemplifies minimalist design that removes technical barriers. No axes, labels, or numbers—just color conveying 170 years of temperature change.
Why It Works
By stripping away everything except the essential message (warming trend), Warming Stripes communicates climate change to non-scientists who would struggle with traditional scientific visualizations. Sometimes less truly is more.
Exercise 11.1: Data Feminism Application
Create a visualization that explicitly addresses one Data Feminism principle: either (a) make data collection labor visible, (b) challenge a binary categorization in existing data, or (c) design for emotional engagement while maintaining accuracy.
Related Resources
Decolonize Dev
Essential Reading
Check Your Understanding
Test your comprehension of the key concepts from this module.
Choose one advanced technique (small multiples, animation, network visualization, uncertainty) and identify a dataset or question from your work where it would be valuable. Sketch how you would implement it.
Capstone Project Portfolio Development
Complete an end-to-end visualization project from data acquisition through published output. Develop a professional portfolio and give structured critique.
Capstone Timeline
The capstone project spans 3-4 weeks with milestone submissions at each stage.
Week 1: Proposal
Identify dataset, define audience and purpose, sketch 3+ visualization approaches.
Week 2: Prototype
Build functional draft visualizations, document design decisions, peer review exchange.
Week 3: Refinement
Incorporate feedback, polish design, prepare presentation materials.
Week 4: Presentation
Present to class and external reviewers; publish to portfolio.
Portfolio Platforms
| Platform | Best For | Considerations |
|---|---|---|
| GitHub Pages | Code-focused, version control, free hosting | Requires basic HTML/CSS; shows technical credibility |
| Observable | JavaScript/D3 notebooks, interactive, community | Great for web interactivity; less traditional layout |
| Tableau Public | Dashboard hosting, industry recognition | All work is public; limited customization |
| Personal Website | Full control, multimedia integration | Requires maintenance; shows professionalism |
Structured Critique Framework
Effective critique follows a structured process, not "I like/don't like" reactions.
1. Description
What do you see? Describe without judgment. "I see a bar chart with 12 categories..."
2. Analysis
How does it work? What design choices were made? "Color encodes region..."
3. Interpretation
What message does it convey? "This suggests that coverage varies significantly..."
4. Evaluation
How effective? What could improve? "The legend placement could be clarified..."
Published portfolio containing capstone project plus 2-3 additional polished visualizations from course exercises. Portfolio includes: project descriptions, design rationales, documented process, and code/data sources where applicable.
Portfolio Inspiration & Resources
Check Your Understanding
Test your comprehension of the key concepts from this module.
Outline a capstone visualization project: What question will you explore? What data sources will you use? What audience are you designing for? What mix of chart types and techniques will you employ?
Chart Chooser Find the Right Chart
Answer a few questions about your data and purpose to get evidence-based chart recommendations.
Chart Chooser Tool
Select your data type and visualization goal to receive personalized recommendations.
Color Palette Lab Test & Compare
Explore colorblind-safe palettes and test them with simulated color vision deficiencies.
Recommended Palettes
These palettes are designed to be perceptually uniform, colorblind-safe, and print-friendly.
Viridis (Sequential)
Best for: ordered data, heatmaps, choropleths
ColorBrewer Blues (Sequential)
Best for: single-hue progression, neutral aesthetic
RdBu (Diverging)
Best for: data with meaningful midpoint (change, deviation)
Set2 (Categorical)
Best for: unordered categories (max 6-8 distinct colors)
Always Test Your Palettes
Use ColorBrewer for palette selection and Color Oracle or Coblis to simulate colorblind perception.
Visualization Glossary Key Terms
Essential terminology for data visualization, organized by category.
Foundational Concepts
| Term | Definition |
|---|---|
| Data-Ink Ratio | The proportion of a graphic's ink devoted to non-redundant data display. Coined by Edward Tufte; higher ratios indicate more efficient visualizations. |
| Chartjunk | Visual elements that do not convey data-information—decorative elements, heavy gridlines, unnecessary 3D effects. |
| Preattentive Attributes | Visual properties processed unconsciously in ~200ms: color, size, orientation, position. Used to direct attention strategically. |
| Small Multiples | Repeating the same chart structure across panels to compare categories, time periods, or conditions. |
| Grammar of Graphics | Leland Wilkinson's framework decomposing visualizations into components: data, aesthetics, geometries, scales, coordinates, facets. |
Chart Types
| Term | Definition |
|---|---|
| Choropleth Map | Geographic regions filled with color to represent data values. Caution: large areas visually dominate regardless of importance. |
| Dot Plot | Points positioned along a scale. Often superior to bar charts for precise comparison without baseline distortion. |
| Slope Chart | Lines connecting two time points showing change. Excellent for before/after comparisons across categories. |
| Sparkline | Intense, word-sized graphics embedded within text. Shows trends without breaking reading flow. |
| Treemap | Hierarchical data shown as nested rectangles. Size encodes value; nesting encodes hierarchy. |
| Sankey Diagram | Flow diagram where width of connections represents quantity. Used for showing flows, transfers, or transformations. |
Color & Perception
| Term | Definition |
|---|---|
| Sequential Palette | Colors progressing from light to dark for ordered data without midpoint. Example: population density. |
| Diverging Palette | Two hues diverging from neutral midpoint. For data with meaningful center: change from baseline, deviation from target. |
| Categorical Palette | Distinct hues for unordered groups. Maximum 7-8 distinguishable colors before confusion. |
| Perceptually Uniform | Equal numerical steps produce equal perceptual steps. Viridis palette achieves this; rainbow does not. |
| Deuteranopia | Most common color vision deficiency (~6% of males). Cannot distinguish red-green. Design must account for this. |
Data Visualization Lexicon
Essential terminology for data visualization practitioners, from perceptual principles to technical implementation.
Data-Ink Ratio
Edward Tufte's principle: maximize the share of ink devoted to data, minimize non-data ink.
Pre-attentive Processing
Visual features processed before conscious attention: color, size, orientation, motion.
Small Multiples
Series of similar charts showing different slices of data, enabling comparison through consistent design.
Lie Factor
Ratio of visual effect size to data effect size. A lie factor >1.05 or <0.95 indicates distortion.
Reflection
Find a data visualization from a recent development report. Evaluate it using the lexicon terms: What's the data-ink ratio? Are pre-attentive features used effectively? Is there any lie factor present?
Video Lectures
Curated video resources from leading data visualization practitioners and researchers.
Hans Rosling's TED Talks
The master of animated data storytelling. Essential viewing for anyone visualizing development data.
Watch on TED →
Supplementary Video Resources
Additional curated video lectures on data visualization principles and practice.
Coming Soon
Meet the Founders of ImpactMojo
This course is brought to you by two practitioners passionate about democratizing development education.
Varna
Founder & Lead of Learning Design
Development Economist with a PhD, specializing in social impact measurement, gender studies, and development research across South Asia.
Vandana
Co-Founder & Lead of Partnerships
Education professional with 15+ years designing impactful learning programs and driving large-scale education initiatives across India.