TidyTuesday Week 44: Lead Levels in the Drinking Water of Flint Michigan

This week we are exploring lead levels in water samples collected in Flint, Michigan in 2015. The data comes from a paper by Loux and Gibson (2018) who advocate for using this data as a teaching example in introductory statistics courses.

TidyTuesday
Data Visualization
Python Programming
2025
Author

Peter Gray

Published

November 6, 2025

Chart Violin Plots of lead concentration in FLint Mchigan

1. Python code

Show code 
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

# Beide Datensätze laden
flint_mdeq = pd.read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-11-04/flint_mdeq.csv')


flint_vt = pd.read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-11-04/flint_vt.csv')

# MDEQ Daten bereinigen
flint_mdeq = flint_mdeq[flint_mdeq["notes"].isna()]
flint_mdeq["average"] = flint_mdeq[['lead', 'lead2']].mean(axis=1)
flint_mdeq = flint_mdeq.dropna(subset=["average"])

# VT Daten bereinigen (nur 'lead' Spalte verwenden)
flint_vt = flint_vt.dropna(subset=["lead"])



# Daten für Violin Plot vorbereiten
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 6))

# MDEQ Violin Plot
violin1 = ax1.violinplot([flint_mdeq["average"]], positions=[1], 
                        showmeans=True, showmedians=True, showextrema=True)
for pc in violin1['bodies']:
    pc.set_facecolor('#87CEEB')
    pc.set_alpha(0.7)

ax1.boxplot([flint_mdeq["average"]], positions=[1], widths=0.1, 
           patch_artist=True, boxprops=dict(facecolor='white', alpha=0.8))
ax1.axhline(y=15, color='red', linestyle='--', alpha=0.7, label='EPA Action Level (15 ppb)')
ax1.set_title('MDEQ Dataset\n(Mean Lead Concentration)', fontweight='bold', fontsize =10)
ax1.set_ylabel('Lead Concentration (ppb)')
ax1.grid(False)
ax1.set_xticks([1])
ax1.set_xticklabels(['MDEQ Dataset'])
ax1.legend(loc='lower left', bbox_to_anchor=(-0.1, -0.25))

# VT Violin Plot
violin2 = ax2.violinplot([flint_vt["lead"]], positions=[1], 
                        showmeans=True, showmedians=True, showextrema=True)
for pc in violin2['bodies']:
    pc.set_facecolor('#98FB98')
    pc.set_alpha(0.7)

ax2.boxplot([flint_vt["lead"]], positions=[1], widths=0.1, 
           patch_artist=True, boxprops=dict(facecolor='white', alpha=0.8))
ax2.axhline(y=15, color='red', linestyle='--', alpha=0.7, label='EPA Action Level (15 ppb)')
ax2.set_title('Viginia Tech \n Dataset', fontweight='bold', fontsize = 10)
ax2.set_ylabel('Lead Concentration (ppb)')
ax2.set_xticks([1])
ax2.set_xticklabels(['VT'])
ax2.grid(False)  # Removes grid lines
ax2.legend().remove()
plt.subplots_adjust(wspace = 0.4)
plt.tight_layout()

Back to top