Topic 7.2: The Right Chart for the Story

A professional journalist never uses the same format for every story. A national election result needs a bar chart; a stock price over a year needs a line chart; the relationship between two variables needs a scatter plot; the distribution of responses needs a histogram. Each format is designed to answer a specific type of question.

The same principle applies to data visualization. Choosing the wrong chart type doesn't make the data wrong — it makes it misleading or confusing. The right chart makes the answer to your question obvious within two seconds of looking at it.

import matplotlib.pyplot as plt
import pandas as pd

# Load student results dataset
df = pd.read_csv('../Datasets/7_2_student_results.csv')

print("Student Results Dataset:")
print(df.head(10))
print(f"\nShape: {df.shape}")

Use a bar chart when you want to compare discrete, named categories. The question it answers is: Which category is largest? Which is smallest? The bars make differences in magnitude visually obvious.

In the student dataset, we have three subjects: Math, Science, and English. To compare average performance across subjects, a bar chart is the natural choice. Each subject is a category; the average score is the value.

# Bar Chart — compare average scores by subject
subjects = ['math_score', 'science_score', 'english_score']
averages = [df[s].mean() for s in subjects]
labels = ['Math', 'Science', 'English']

fig, ax = plt.subplots(figsize=(8, 5))
ax.bar(labels, averages, color='steelblue')

ax.set_title('Average Subject Scores')
ax.set_xlabel('Subject')
ax.set_ylabel('Average Score')
ax.set_ylim(0, 100)

plt.tight_layout()
plt.show()

The code works as follows: first, we build a list of column names (subjects) and compute the mean of each using a list comprehension — [df[s].mean() for s in subjects] iterates over each column name and returns the average value. ax.bar(categories, values) draws one vertical bar for each category, with height equal to the corresponding value. The ax.set_ylim(0, 100) call forces the y-axis to start at zero. This is important for bar charts: starting the y-axis above zero exaggerates differences and can mislead the reader.

Use a line chart when your data follows a natural sequence — most commonly time. The question it answers is: How does this value change across a series of ordered points? The connected line reveals the direction and shape of the trend at a glance.

Use a line chart when your data has a natural order and you want to see how a value changes across that sequence. Here we use the same student dataset — plotting math scores for each student in order. The connected line makes it easy to spot which students score high or low and how scores vary across the class.

# Line Chart — math score for each student in sequence
fig, ax = plt.subplots(figsize=(10, 5))
ax.plot(df['name'], df['math_score'])

ax.set_title('Math Score by Student')
ax.set_xlabel('Student')
ax.set_ylabel('Math Score')
plt.xticks(rotation=45)

plt.tight_layout()
plt.show()

Use a scatter plot when you want to show the relationship between two continuous variables. The question it answers is: When variable A increases, does variable B also increase? Decrease? Show no pattern?

For the student dataset, a natural question is: do students who study more hours get higher math scores? Each student becomes a dot — their study hours on the x-axis, their math score on the y-axis. If a correlation exists, the dots will form a visible diagonal pattern.

# Scatter Plot — correlation between study hours and math score
fig, ax = plt.subplots(figsize=(8, 6))
ax.scatter(df['study_hours'], df['math_score'],
           color='purple', alpha=0.7)

ax.set_title('Study Hours vs Math Score')
ax.set_xlabel('Study Hours per Day')
ax.set_ylabel('Math Score')

plt.tight_layout()
plt.show()

ax.scatter(x, y) draws one dot for each data point, placing it at the x-coordinate (study hours) and y-coordinate (math score) for that student. If a positive trend exists, dots will slope upward from left to right. With 30 students, some dots may overlap — you can address this with the alpha parameter once you start customizing.

Use a histogram when you want to understand how a single variable is distributed. The question it answers is: Where do most values cluster? Is the data symmetric, skewed, or spread out?

A histogram divides the full range of values into equal-width bins and counts how many data points fall into each bin. The height of each bar represents that count. Unlike a bar chart (which compares named categories), a histogram shows the frequency landscape of a continuous variable.

# Histogram — distribution of math scores
fig, ax = plt.subplots(figsize=(8, 5))
ax.hist(df['math_score'], bins=8, color='steelblue', edgecolor='white')

ax.set_title('Distribution of Math Scores')
ax.set_xlabel('Math Score')
ax.set_ylabel('Number of Students')

plt.tight_layout()
plt.show()

ax.hist(data, bins=8) automatically divides the full range of math_score values into 8 equal-width intervals (bins) and counts how many students fall into each one. The height of each bar is that count. You do not need to group the data manually — Matplotlib handles all the counting for you.

The bins parameter controls how many equal-width intervals the range is divided into. Too few bins (e.g., 2) loses all the detail. Too many bins (e.g., 30 for 30 data points) makes every bar height 0 or 1, which is equally useless. A good starting point for small datasets is 8–12 bins. The edgecolor='white' parameter adds a thin white border between bars for visual clarity.

Every chart type is built to answer a specific kind of question. Before reaching for a chart, ask: What is my question? The answer determines the chart.