Technical Deep-Dive

    Data Visualization Best Practices

    Create clear, effective figures and tables for your dissertation. From chart selection to APA formatting—elevate your visual communication.

    Design principles
    Table formatting
    APA 7th edition
    "Above all else, show the data."
    — Edward Tufte

    The goal of every visualization is to communicate data clearly. Everything else—colors, gridlines, decorations—should serve that purpose or be removed.

    Choosing the Right Chart Type

    Match your visualization to your data and message. Each chart type serves specific purposes.

    Bar Charts

    Comparing categories or groups

    Examples:

    • • Comparing means across groups
    • • Frequency distributions
    • • Likert scale responses

    Tips:

    • Start y-axis at zero for honest comparison
    • Order bars logically (alphabetically, by value, or conceptually)
    • Use horizontal bars for many categories or long labels

    Line Charts

    Showing trends over time or continuous relationships

    Examples:

    • • Longitudinal data
    • • Time series
    • • Pre-post comparisons across groups

    Tips:

    • Reserve for truly continuous data or time series
    • Don't connect categorical data with lines
    • Limit to 4-5 lines for readability

    Scatter Plots

    Showing relationships between two continuous variables

    Examples:

    • • Correlation visualization
    • • Regression analysis
    • • Identifying outliers

    Tips:

    • Add regression line when appropriate
    • Use jittering for overlapping points
    • Consider adding marginal distributions

    Pie/Donut Charts

    Showing parts of a whole (use sparingly)

    Examples:

    • • Demographic composition
    • • Budget allocation
    • • Simple proportions

    Tips:

    • Use only for parts that sum to 100%
    • Limit to 5-6 slices maximum
    • Consider bar charts for easier comparison
    • Start at 12 o'clock, largest slice first

    Box Plots

    Showing distribution characteristics and outliers

    Examples:

    • • Comparing distributions across groups
    • • Identifying outliers
    • • Showing variance

    Tips:

    • Excellent for comparing multiple groups
    • Always explain components to readers
    • Consider violin plots for more detail

    Heat Maps

    Showing patterns in matrices or correlations

    Examples:

    • • Correlation matrices
    • • Cross-tabulations
    • • Temporal patterns

    Tips:

    • Use colorblind-friendly palettes
    • Order rows/columns meaningfully
    • Include numerical values for precision

    Table Design Principles

    Well-designed tables communicate complex information efficiently. Follow these principles for clarity.

    Purpose

    Every table should answer a specific question. If it doesn't serve analysis or support arguments, remove it.

    Example: Bad: 'Table of all variables'. Good: 'Descriptive Statistics for Key Study Variables'

    Structure

    Organize rows and columns logically. Group related items. Use clear, concise headers.

    Example: Put dependent variables in rows, groups/conditions in columns for easy comparison

    Formatting

    Align numbers by decimal point. Use consistent decimal places. Minimize lines and borders.

    Example: Use 2 decimal places for most statistics, 3 for p-values. No vertical lines.

    Labeling

    Include descriptive title, clear labels, units of measurement, and notes for abbreviations.

    Example: Note: *p < .05, **p < .01, ***p < .001. CI = confidence interval.

    APA 7th Edition Formatting

    Follow these formatting requirements for figures and tables in APA style.

    Figure number
    Bold, flush left above figure (e.g., 'Figure 1')
    Figure title
    Italic, title case, double-spaced below number
    Image
    Clear, high-resolution, black and white for print
    Legend
    Within figure boundaries, explaining symbols/colors
    Note
    Below figure, explains abbreviations, significance levels

    Common Visualization Mistakes

    Using 3D effects on charts

    3D distorts proportions and hinders comparison. Always use 2D charts—they're clearer and more honest.

    Truncating the y-axis to exaggerate differences

    Start bar charts at zero. If differences are meaningful, they'll be visible. Break the axis only when necessary and clearly indicate the break.

    Using pie charts for comparisons

    Humans compare lengths better than angles. Use bar charts when comparing categories. Reserve pie charts for simple part-to-whole relationships.

    Too much color without purpose

    Use color to encode meaning (groups, values, emphasis). Gray is your friend—use it for secondary information.

    Insufficient labels or missing legends

    Every axis needs a label with units. Every figure needs a clear legend. Readers shouldn't have to guess what symbols mean.

    Low-resolution or pixelated images

    Export at 300+ DPI for print. Use vector formats (PDF, SVG) when possible. Test figures at actual print size.

    Frequently Asked Questions

    Need Help with Your Data Presentation?

    Our data analysis experts can help you create publication-quality figures and tables that clearly communicate your research findings.