# Data Visualization

Psychology I · Statistics · https://tryals.app/learn/psychology-i/data-visualization

## Chart Selection

Different data calls for different pictures. Using the wrong one can mislead.

A **histogram** shows the distribution of a single **continuous** variable, using touching bins to reveal shape. A **bar chart** compares distinct **categories** with gaps between bars.

A **scatterplot** maps the relationship between two continuous variables. A **box plot** summarizes the **five-number summary** (minimum, Q1, median, Q3, maximum) and spots outliers.

| Chart | Best For |
|---|---|
| Histogram | Distribution shape |
| Bar chart | Category comparison |
| Scatterplot | Two-variable relationship |
| Box plot | Five-number summary |
| Line graph | Trends over time |

## Visual Traps

Visualizations can deceive even with accurate data. A **truncated y-axis** starts above zero, making small differences look dramatic.

Other tricks include inconsistent scaling, cherry-picked time windows, and 3D effects that distort volume perception.

> **Common pitfall:** Using a bar chart with a truncated axis to compare similar values. A bar chart is only honestly interpretable when its vertical axis begins at zero.

Always inspect where a chart's axis starts before trusting its visual impact.

## Practice questions

4 of this lesson's 10 practice questions, with answers. The full set is in the app.

### 1. Sort each graphing practice as honest or misleading.

**Answer:**

- Honest practice: Starting a bar chart's y-axis at zero, Choosing a scatterplot to show two continuous variables
- Misleading technique: Truncating the y-axis to exaggerate a small difference, Using 3D effects that distort area and volume perception, Cherry-picking a narrow time window that hides the overall trend

**Why:** Zero-based axes and matching the chart to the data are honest; truncated axes, distorting 3D, and cherry-picked windows all inflate a pattern beyond the actual numbers. The distinction is faithfulness to the data, not visual impact.

Page: https://tryals.app/practice/psychology-i/data-visualization/sort-each-graphing-practice-as-honest-or-misleading

### 2. Tufte argued a chart should maximise the **data-ink ratio**: the share of a graphic that encodes actual numbers. Why does decoration cost more than aesthetics?

A. Because journals require plain charts, making decoration a submission risk
B. Because decoration is always used deliberately to mislead
C. Because decorated charts take longer to render and print
D. Because every non-data mark competes with the data for attention

**Answer:** D. Because every non-data mark competes with the data for attention

**Why:** Ink that encodes nothing still gets looked at, and a 3D pie chart actively distorts, perspective makes near slices read larger. The reason to strip a chart is not taste but bandwidth: everything removed is attention returned to the comparison you wanted the reader to make.

Page: https://tryals.app/practice/psychology-i/data-visualization/tufte-argued-a-chart-should-maximise-the-data-ink-ratio-the-share-of

### 3. A box plot displays the five-number summary. Order those five values from lowest to highest.

**Answer:**

1. Minimum
2. First quartile (Q1)
3. Median
4. Third quartile (Q3)
5. Maximum

**Why:** Minimum, Q1, median, Q3, maximum - ascending. The box spans Q1 to Q3 (the middle 50%), the line inside marks the median, and the whiskers reach toward the extremes.

Page: https://tryals.app/practice/psychology-i/data-visualization/a-box-plot-displays-the-five-number-summary-order-those-five-values

### 4. Select every statement that is TRUE about data visualization.

A. A scatterplot is the natural chart for two continuous variables.
B. A bar chart is ideal for showing the distribution shape of one continuous variable.
C. A histogram's touching bars signal a continuous variable.
D. Starting a bar chart's axis above zero makes the comparison more honest.
E. A truncated y-axis can exaggerate a genuinely small difference.

**Answer:** A. A scatterplot is the natural chart for two continuous variables.; C. A histogram's touching bars signal a continuous variable.; E. A truncated y-axis can exaggerate a genuinely small difference.

**Why:** Touching bars, truncated-axis distortion, and scatterplots for two continuous variables are all true. A histogram (not a bar chart) shows one variable's distribution, and starting the axis above zero makes a bar chart LESS honest.

Page: https://tryals.app/practice/psychology-i/data-visualization/select-every-statement-that-is-true-about-data-visualization