Stacked bar charts in Matplotlib - Step-by-Step Execution

Concept Flow - Stacked bar charts

Prepare data for categories

↓

Plot first bar segment

↓

Plot next bar segment on top

↓

Repeat for all segments

↓

Display stacked bar chart

Stacked bar charts build bars by stacking segments for each category one on top of another, showing parts of a whole.

Execution Sample

Matplotlib

import matplotlib.pyplot as plt
categories = ['A', 'B', 'C']
values1 = [3, 5, 1]
values2 = [2, 3, 4]
plt.bar(categories, values1)
plt.bar(categories, values2, bottom=values1)
plt.show()

This code draws a stacked bar chart with two segments per category.

Execution Table

Step	Action	Data Used	Bottom Parameter	Bar Height Plotted	Visual Effect
1	Plot first bar segment	[3, 5, 1]	None	[3, 5, 1]	Bars of height 3,5,1 appear for A,B,C
2	Plot second bar segment	[2, 3, 4]	[3, 5, 1]	[2, 3, 4]	Bars stacked on top, total heights 5,8,5
3	Display chart	N/A	N/A	N/A	Stacked bar chart shown with two layers per category

💡 All segments plotted and chart displayed

Variable Tracker

Variable	Start	After Step 1	After Step 2	Final
categories	['A', 'B', 'C']	['A', 'B', 'C']	['A', 'B', 'C']	['A', 'B', 'C']
values1	[3, 5, 1]	[3, 5, 1]	[3, 5, 1]	[3, 5, 1]
values2	[2, 3, 4]	[2, 3, 4]	[2, 3, 4]	[2, 3, 4]
bottom	None	None	[3, 5, 1]	[3, 5, 1]

Key Moments - 3 Insights

Why do we use the 'bottom' parameter when plotting the second bar segment?

The 'bottom' parameter tells matplotlib where to start the next bar segment vertically, so it stacks on top of the first segment (see execution_table step 2). Without it, bars would overlap from zero.

What happens if the 'bottom' parameter is not the same length as the data?

Matplotlib will raise an error because it needs matching lengths to stack bars correctly. Each bar segment must align with the previous segment's height (refer to variable_tracker for lengths).

Can we stack more than two segments?

Yes! You keep adding bars with 'bottom' equal to the sum of all previous segments for each category, stacking them higher (concept_flow shows repeated stacking).

Visual Quiz - 3 Questions

Look at the execution_table, what is the 'bottom' parameter value when plotting the second bar segment?

A None

B [2, 3, 4]

C [3, 5, 1]

D [5, 8, 5]

💡 Hint

Check the 'Bottom Parameter' column in execution_table row 2

At which step do the bars first appear stacked on top of each other?

A Step 1

B Step 2

C Step 3

D Never stacked

💡 Hint

Look at the 'Visual Effect' column in execution_table rows 1 and 2

If we add a third segment with values [1, 2, 3], what should the 'bottom' parameter be for correct stacking?

A [5, 8, 5]

B [2, 3, 4]

C [3, 5, 1]

D [1, 2, 3]

💡 Hint

Sum the first two segments from variable_tracker values1 and values2

Concept Snapshot

Stacked bar charts show parts of categories stacked vertically.
Use plt.bar() for each segment.
Set 'bottom' to previous segments' heights.
Repeat for all segments.
Result: bars stacked to show totals and parts.

Full Transcript

Stacked bar charts build bars by stacking segments for each category one on top of another. First, we plot the base segment with plt.bar(). Then, for each next segment, we plot bars with the 'bottom' parameter set to the sum of all previous segments for that category. This stacks the bars visually. The example code plots two segments for categories A, B, and C. The execution table shows step-by-step how the bars are drawn and stacked. The variable tracker shows how data and the 'bottom' parameter change. Key moments clarify why 'bottom' is needed and how stacking works. The quiz tests understanding of stacking steps and parameters.

Practice

1. What is the main purpose of using a stacked bar chart in data visualization?

easy

A. To create a pie chart with multiple slices

B. To display data points as individual dots on a graph

C. To plot a line connecting data points over time

D. To show how different parts contribute to the total within each group

Solution

1. Step 1: Understand the concept of stacked bar charts

   Stacked bar charts combine multiple data parts into one bar to show totals and parts.

2. Step 2: Identify the purpose of stacking

   Stacking helps visualize how each part contributes to the total for each group.

3. Final Answer:

   To show how different parts contribute to the total within each group -> Option D

4. Quick Check:

   Stacked bars = parts + total [OK]

Hint: Stacked bars add parts vertically to show totals [OK]

Common Mistakes:

• Confusing stacked bars with line charts
• Thinking stacked bars show individual points only
• Mixing stacked bars with pie charts

2. Which matplotlib parameter is used to stack bars on top of each other in a bar chart?

easy

A. height

B. bottom

C. width

D. align

Solution

1. Step 1: Recall bar chart parameters

   In matplotlib, height sets bar size, width sets bar thickness, align controls bar alignment.

2. Step 2: Identify stacking parameter

   The bottom parameter sets where the bar starts vertically, allowing stacking on previous bars.

3. Final Answer:

   bottom -> Option B

4. Quick Check:

   Stack bars use bottom parameter [OK]

Hint: Use bottom to place bars on top of others [OK]

Common Mistakes:

• Using height to stack bars (it sets bar size)
• Confusing width with stacking
• Using align which only shifts bars horizontally

3. What will be the output of this code snippet?

import matplotlib.pyplot as plt

labels = ['A', 'B']
values1 = [3, 5]
values2 = [2, 4]

plt.bar(labels, values1)
plt.bar(labels, values2, bottom=values1)
plt.show()

medium

A. Two bars per label, stacked vertically

B. Error because bottom parameter is incorrect

C. Only one bar per label showing values2

D. Two bars per label, side by side

Solution

1. Step 1: Analyze the first bar call

   The first plt.bar draws bars of height 3 and 5 for labels 'A' and 'B'.

2. Step 2: Analyze the second bar call with bottom

   The second plt.bar draws bars of height 2 and 4 stacked on top of the first bars using bottom=values1.

3. Final Answer:

   Two bars per label, stacked vertically -> Option A

4. Quick Check:

   bottom stacks bars vertically [OK]

Hint: bottom stacks bars vertically on previous bars [OK]

Common Mistakes:

• Thinking bars appear side by side without bottom
• Assuming only one bar per label is drawn
• Believing bottom parameter causes error

4. Identify the error in this code for creating a stacked bar chart:

import matplotlib.pyplot as plt

labels = ['X', 'Y']
values1 = [4, 6]
values2 = [3, 5]

plt.bar(labels, values1)
plt.bar(labels, values2, bottom=values2)
plt.show()

medium

A. Labels list is missing an element

B. plt.bar cannot be called twice

C. Using bottom=values2 instead of bottom=values1 causes incorrect stacking

D. values1 and values2 must be the same length

Solution

1. Step 1: Check the bottom parameter usage

   The second bar uses bottom=values2, which stacks bars on themselves, not on the first bars.

2. Step 2: Correct stacking requires bottom=values1

   To stack bars correctly, the second bars should start at the height of the first bars, so bottom=values1 is needed.

3. Final Answer:

   Using bottom=values2 instead of bottom=values1 causes incorrect stacking -> Option C

4. Quick Check:

   bottom must be previous bar heights [OK]

Hint: Stack bars using bottom=previous bar heights [OK]

Common Mistakes:

• Setting bottom to current bar heights instead of previous
• Thinking plt.bar can only be called once
• Ignoring length matching of values

5. You have three categories of sales data for two stores:

store_labels = ['Store1', 'Store2']
sales_A = [10, 15]
sales_B = [5, 7]
sales_C = [3, 4]

How do you create a stacked bar chart showing all three sales categories stacked for each store?

hard

A. Plot sales_A, then sales_B with bottom=sales_A, then sales_C with bottom=sales_A + sales_B

B. Plot sales_A, sales_B, sales_C all with bottom=0

C. Plot sales_A, then sales_B with bottom=sales_A, then sales_C with bottom=sales_B

D. Plot sales_A with bottom=sales_B, sales_B with bottom=sales_C, sales_C with bottom=sales_A

Solution

1. Step 1: Plot the first category normally

   Plot sales_A bars starting at zero (default).

2. Step 2: Stack second category on first

   Plot sales_B bars with bottom=sales_A to stack on top of sales_A.

3. Step 3: Stack third category on sum of first two

   Plot sales_C bars with bottom=[a+b for a,b in zip(sales_A, sales_B)] to stack on top of both.

4. Final Answer:

   Plot sales_A, then sales_B with bottom=sales_A, then sales_C with bottom=sales_A + sales_B -> Option A

5. Quick Check:

   Stack each bar on sum of all below [OK]

Hint: Stack bars using cumulative sums for bottom [OK]

Common Mistakes:

• Stacking third category only on second, ignoring first
• Plotting all bars at bottom=0 causing overlap
• Using wrong bottom values causing incorrect stacking