EvalCompassEvalCompass
Analysis & synthesisChart selectionEvaluation reportingAccessibilityEvergreenCommunication

Data Visualization

Data Visualization
Also known as: Data visualization、Data viz、Evaluation report graphics

Communicates numerical comparisons, trends and distributions visually. Select charts suited to questions and show axes, units, denominators and sources to reduce misreading. [1] [2]

Data visualization chart selection checklist — openReferences and checked sections
Difficulty
Introductory
Cost
Low
Duration
Several hours–several days for data preparation, design, charting and revision
Content updated: 2026-10-07Source checked: 2026-09-13 (includes AI cross-checking)No documented human reviewIncludes fictional examples

How it works at a glance

Bar charts encode quantities through length, so start their baseline at 0. Align scales between compared charts and choose appropriate time-series or distribution displays. [1] [2]

Do not convey meaning through color alone: add labels, explanations and data tables where needed. Show uncertainty and missingness, and check readability across screen sizes. [1] [2]

  1. Clarify the message
  2. Select chart types
  3. Optimize data ink
  4. Design color and accessibility
  5. Add uncertainty and notes
  6. Review and test
Sequence of implementation steps. See “Implementation steps” for details of each stage.

When to use it and when it is unsuitable

When to use it

  • When making statistics intuitively understandable in reports or presentations
  • When comparing groups or patterns over time
  • When understanding or explaining spatial or categorical distributions by area or characteristics
  • When communicating to readers with varied statistical literacy, including policy-makers and the public

When it is unsuitable

  • When data quality or reliability is low. Attractive graphics conceal uncertainty; disclose it first.
  • Consider tables when reading a few exact values is the priority and a chart is less clear.
  • When software or skills are inadequate and misleading graphics may result

Participants, roles, and deliverables

Participants and roles

1–2 evaluators skilled in Excel, R, Python, Tableau or Power BI. For large reports, collaboration with designers is desirable; consider outsourcing.

Main deliverables

  • Report and presentation chart set
  • Chart selection rationale
  • Accessibility checklist: color vision diversity and contrast checked

Implementation steps

  1. Clarify the message
    Define the question and what data support. Titles should specify observed differences, not claim effects or significance without causal analysis or tests.
  2. Select chart types
    Match charts to comparisons, composition, relationships, distributions or trends. Consider bars or tables for fine percentage comparisons and pies for composition; do not decide by element count alone.
  3. Optimize data ink
    Reduce decoration while preserving axes, ticks, units and legends needed for comparisons. Combine visual simplicity with accurate reading.
  4. Design color and accessibility
    Combine labels, line styles, shapes or tables rather than color alone. Check applicable contrast standards for text and graphics separately. Provide alternative text or source tables.
  5. Add uncertainty and notes
    Show confidence intervals, standard errors or sample sizes with error bars or annotations. Add definitions, methods, collection year and exclusions below the chart.
  6. Review and test
    Ask intended readers what they infer and how they understand denominators and differences. Correct unreadable labels and misconceptions; do not judge solely by fixed viewing times.

Strengths and limitations

Strengths

  • Communicates numerical patterns through charts appropriate to comparisons, distributions or trends.
  • Patterns, outliers and trends are immediately visible and readily shared.
  • For multilingual use, check units, labels and conventions and explain interpretation.
  • Interactive tools such as Tableau or Power BI enable readers to explore data.

Limitations

  • Misleading axes for length- or area-based quantities or inconsistent comparison scales distort perceptions. Set and disclose line- or scatterplot ranges appropriate to context and variation.
  • Wrong chart choices, such as lines for categorical data, mislead.
  • Visuals readily show what happens, but causal explanations of why require accompanying prose.
  • Interactive implementation requires technical skills such as HTML or JavaScript.

Common pitfalls

  • Nonzero bar-chart baselines exaggerate small differences. Always start bars at 0; line-chart ranges may vary with context.
  • Making composition difficult to compare. Test bars, pies or tables with readers according to subjects and precision.
  • Decorative chartjunk: 3D, shadows, excess colors and background images hinder reading. Simplicity expresses the greatest honesty in evaluation reporting.

Conditions of use and evidence

  • Use axes that do not distort bar lengths and communicate units and chart content in prose too. [1] [2]
Fictional example; not actual results

Show application rate differences accurately

A fictional illustrative design comparing area A at 50% and B at 60% in a bar chart.

Display elementDesign adoptedReason
Vertical axis0〜100%Bar length accurately shows the 10-percentage-point difference.
LabelA: 50/100 people; B: 60/100 peopleShow denominators and percentages.
TextB is 10 percentage points higher than A.Content remains readable without color.

Common errors in judgment

Common mistake

Truncating bar axes to exaggerate gaps

How to revise

Use a zero baseline and note the difference.

ReasonBar lengths otherwise misrepresent quantity ratios.

Common mistake

Distinguishing series only through a color legend

How to revise

Add direct labels, line styles, tables or alternative descriptions.

ReasonPeople with color vision difficulties and screen-reader users can also access information.

Organize your thinking by writing

Data visualization chart selection checklist

A practical template is available to fill in and print on A4 paper.

Open worksheet

Related practical guides

How to write an evaluation report

Related methods

Descriptive Statistics and Cross-tabulationEvaluation Report WritingMixed Methods ResearchEvaluation Dissemination and CommunicationPerformance Measurement

References and checked sections

  1. Axes and gridlines
    Office for National Statistics
    Section consulted
    Axis start point / Consistent axis scales
    How it relates to the text
    Use axes that do not distort bar lengths and communicate units and chart content in prose too.
    Source checked: (includes AI cross-checking)
  2. Chart text
    Office for National Statistics
    Section consulted
    Chart alternative text / Sources / Axis titles
    How it relates to the text
    Use axes that do not distort bar lengths and communicate units and chart content in prose too.
    Source checked: (includes AI cross-checking)

Editorial and checking information

Content updated: Source checked: (includes AI cross-checking)No documented human reviewIncludes fictional examplesEditorial policy and scope of checkingCite this pageReport a correction

How to cite this page

EvalCompass Editorial Team. “Data Visualization”. EvalCompass, updated 2026-10-07. https://evalcompass.jun-nakatani.com/en/methods/data-visualization/
Add the date you accessed the page. Text created by this site is available under CC BY 4.0. Quotations, external materials, and bundled fonts have their own terms of use. When reusing content, credit the source, link to the license, and indicate any changes.

Suggest a correction or improvement to this page

Change history
  • — Checked sources for definitions and applicability conditions, and added fictional examples, reasons for judgments and examples for every worksheet field. Revised unverified claims about history and adoption. Also reviewed existing procedures and cautions, and adjusted fixed participant counts and thresholds and generalizations about institutions. Preserved field positions, types and counts while revising fixed conditions and thresholds in the forms.
  • — Correction based on independent content review (content-check-causal). Restricted zero-baseline cautions to bar- and area-based charts and specified context-sensitive ranges for line and scatterplots.