Comparing several groups in Python

One-way ANOVA, and the rank-based alternative when variances differ.

import numpy as np
from scipy import stats

rng = np.random.default_rng(3)
python = rng.normal(100, 10, 30)
rust = rng.normal(95, 10, 30)
css = rng.normal(80, 10, 30)

print(round(stats.levene(python, rust, css).pvalue, 4))

anova = stats.f_oneway(python, rust, css)
print(round(anova.statistic, 3), f"{anova.pvalue:.2e}")

pairs = [("py-rust", python, rust), ("py-css", python, css)]
for name, a, b in pairs:
    print(name, f"{stats.ttest_ind(a, b).pvalue:.4f}")

How it works

1. f_oneway takes one array per group.
2. A significant F says the groups differ, not which pair.
3. levene checks the equal-variance assumption first.

Keywords and builtins used here

• as
• for
• print
• round

The run, in numbers

Lines

16

Characters to type

460

Tokens

162

Three-star pace

105 tpm

At the three-star pace of 105 tokens a minute, this run takes about 93 seconds.

Type this snippet

Step 4 of 4 in Hypothesis tests, step 7 of 23 in Scientific computing with SciPy.

← Previous Next →