lesson
40 minHonest Charts
The common ways charts mislead, how to fix them, and how to make charts everyone can read.
Objective
By the end of this lesson, you can spot four misleading chart choices and fix them, and make a chart readable for people with low vision or color blindness.
Four ways charts mislead
- Cut-off axis. A bar chart whose y-axis starts at 95 instead of 0 makes a 2 percent difference look huge. Bars should start at zero.
- Cherry-picked time window. Showing only the years that support your point.
- Counts instead of rates. Bigger places look worse on almost everything. (You saw this in Module 5.)
- No source or units. A chart without a source cannot be checked.
Fixing a cut-off axis
import matplotlib.pyplot as plt
groups = ["Group A", "Group B"]
rate = [96, 98]
fig, (left, right) = plt.subplots(1, 2, figsize=(7, 3))
left.bar(groups, rate)
left.set_ylim(95, 99)
left.set_title("Misleading: axis starts at 95")
right.bar(groups, rate)
right.set_ylim(0, 100)
right.set_title("Honest: axis starts at 0")
plt.tight_layout()
plt.show()
Run it in Colab and compare. Same data, very different story.
Charts everyone can read
- Never use color alone to carry meaning. Add labels or patterns.
- Avoid red-green pairs. Many people cannot tell them apart.
- Write a title that states the finding, and label axes with units.
- Add a short text summary under the chart for screen reader users.
A national example
Judgment check
A chart is an argument. If a chart would make a community look worse than the data supports, or would be used to cut services, slow down and check it with someone from that community.
Honest chart checklist
0 of 6 doneReflection
Discussion
Ask. Answer. Build together.
Loading…