Comparing groups

Improving Your Data Visualizations in Python

Nick Strayer

Instructor

What does this mean?

Values generally higher?
Distribution of values wider? Narrower?
Crucial for representing your data

clusters of people standing together

comparing classes with kernel density plots

kernels stacking up to make KDE

pollution_nov = pollution[pollution.month == 10]

sns.kdeplot(pollution_nov[pollution_nov.city == 'Denver'].O3, color = 'red')
sns.kdeplot(pollution_nov[pollution_nov.city != 'Denver'].O3)

Two overlapping KDEs

# Enable rugplot
sns.kdeplot(pollution_nov[pollution_nov.city == 'Denver'].O3, color='red')
sns.rugplot(pollution_nov[pollution_nov.city == 'Denver'].O3, color='red')
sns.kdeplot(pollution_nov[pollution_nov.city != 'Denver'].O3)

Two overlapping KDEs with a rug plot

A lot of overlapping KDEs

abstract basic beeswarm plots around multiple axes

pollution_nov = pollution[pollution.month == 10]

sns.swarmplot(y="city", x="O3", data=pollution_nov, size=4)
plt.xlabel("Ozone (O3)")

Beeswarm plot of o3 pollution

Let's compare!

Improving Your Data Visualizations in Python