Chart Types and When to Use Them

Choosing the right chart type is one of the most important decisions in data visualisation. The wrong chart can mislead, confuse, or simply fail to communicate the insight. This lesson covers the most common chart types, when to use each one, and common mistakes to avoid.

---

A Decision Framework

Ask yourself these questions before choosing a chart:

1. What is the relationship? — Comparison, distribution, composition, or trend?
2. How many variables? — One, two, or more?
3. What type of data? — Categorical, numerical, temporal, geospatial?
4. Who is the audience? — Analysts who want detail vs executives who want the headline?

---

Chart Selection Guide

Purpose	Chart Types
Comparison	Bar chart, grouped bar, dot plot, lollipop chart
Distribution	Histogram, box plot, violin plot, density plot, strip plot
Composition	Stacked bar, pie chart (use sparingly), treemap, waffle chart
Trend over time	Line plot, area chart, sparkline
Relationship	Scatter plot, bubble chart, heatmap
Part of a whole	Pie chart, donut chart, stacked area chart
Ranking	Horizontal bar chart, bump chart
Geospatial	Choropleth map, bubble map, cartogram

---

Comparison Charts

Bar Chart

The most versatile chart for comparing values across categories.

import matplotlib.pyplot as plt

categories = ["Python", "R", "SQL", "Julia", "Scala"]
popularity = [68, 18, 42, 5, 3]

fig, ax = plt.subplots(figsize=(8, 5))
ax.barh(categories, popularity, color="#4C72B0")
ax.set_xlabel("Popularity (%)")
ax.set_title("Programming Language Popularity in Data Science")
ax.spines[["top", "right"]].set_visible(False)

for i, v in enumerate(popularity):
    ax.text(v + 1, i, str(v), va="center")

plt.tight_layout()
plt.show()

Tip: Use horizontal bar charts when category labels are long. Always start the axis at zero for bar charts.

Grouped Bar Chart

Compare multiple measures across the same categories.

import matplotlib.pyplot as plt
import numpy as np

categories = ["Q1", "Q2", "Q3", "Q4"]
product_a = [120, 135, 148, 162]
product_b = [98, 115, 132, 141]

x = np.arange(len(categories))
width = 0.35

fig, ax = plt.subplots(figsize=(8, 5))
ax.bar(x - width/2, product_a, width, label="Product A", color="#4C72B0")
ax.bar(x + width/2, product_b, width, label="Product B", color="#DD8452")
ax.set_xticks(x)
ax.set_xticklabels(categories)
ax.set_ylabel("Revenue (GBP thousands)")
ax.set_title("Quarterly Revenue by Product")
ax.legend()
ax.spines[["top", "right"]].set_visible(False)
plt.tight_layout()
plt.show()

---

Distribution Charts

Histogram

Shows the frequency distribution of a single numerical variable.

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(42)
data = np.random.normal(loc=170, scale=10, size=1000)

fig, ax = plt.subplots(figsize=(8, 5))
ax.hist(data, bins=30, color="#4C72B0", edgecolor="white")
ax.set_xlabel("Height (cm)")
ax.set_ylabel("Frequency")
ax.set_title("Distribution of Heights")
ax.spines[["top", "right"]].set_visible(False)
plt.tight_layout()
plt.show()

Box Plot

Summarises the distribution through quartiles and highlights outliers.

Component	Meaning
Box	Interquartile range (IQR) — Q1 to Q3
Line inside box	Median (Q2)
Whiskers	Extend to the furthest data point within 1.5 * IQR
Dots beyond whiskers	Outliers

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(42)
data = [np.random.normal(50, 10, 200),
        np.random.normal(60, 15, 200),
        np.random.normal(45, 8, 200)]