Learn Visualizing Point Estimates

A pointplot represents an estimate of the central tendency (mean) by the position of scatter plot points and provides an indication of uncertainty using error bars.

Pointplot vs. Barplot

Technically, they show the exact same data. However, a pointplot connects the estimates with a line. This visualizes the slope of change, making it easier to see how a variable evolves from one category to another.

Key Parameters

To make the plot easier to read (especially in black-and-white), you can customize the markers and lines for different groups:

markers: a list of symbols to use for points (e.g., ['o', 'x']);
linestyles: a list of line styles (e.g., ['-'] for solid, ['--'] for dashed);
dodge=True: slightly shifts the points along the axis so they don't overlap, making error bars distinct.

Example

Here is a pointplot showing how the average bill changes throughout the week. Notice how the dashed line helps distinguish "Lunch" from "Dinner" even without color.


              123456789101112131415161718
            
import seaborn as sns
import matplotlib.pyplot as plt

# Load dataset
df = sns.load_dataset('tips')

# Create a styled pointplot
sns.pointplot(
    data=df,
    x='day',
    y='total_bill',
    hue='time',
    markers=['o', '^'],      # Circle and Triangle markers
    linestyles=['-', '--'],  # Solid and Dashed lines
    dodge=True               # Avoid overlap
)

plt.show()

Task

Swipe to start coding

Visualize the tips given on different days to see if there is a trend.

Set the style to 'ticks'. Customize the background color by passing {'axes.facecolor': 'azure'} as the second argument.
Create a pointplot and assign it to variable g:
- Map 'day' to x and 'tip' to y.
- Group by 'sex' using hue.
- Use 'v' (triangle_down) and 'o' (circle) as markers to distinguish genders.
- Use the 'rocket' palette.
- Enable dodge=True to separate the error bars.
- Set capsize to 0.2 to add caps to the error bars.
- Use solid ('-') and dashed ('--') lines for linestyles.
Set the title to 'Tips pointplot' using the g variable.
Display the plot.

Solution

Everything was clear?

Thanks for your feedback!

Section 1. Chapter 14

single

Ask AI

Ask anything or try one of the suggested questions to begin our chat

Swipe to show menu