The US Judges data

Gender and racial diversity among the judges.
Pillow
PyDyTuesday
TidyTuesday
Author

Manish Datt

Published

June 10, 2025

import pandas as pd
judges_appointments = pd.read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-06-10/judges_appointments.csv')
judges_people = pd.read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-06-10/judges_people.csv')
judges_appointments
judge_id court_name court_type president_name president_party nomination_date predecessor_last_name predecessor_first_name senate_confirmation_date commission_date chief_judge_begin chief_judge_end retirement_from_active_service termination_date termination_reason
0 3419 U. S. District Court, Southern District of New... USDC Barack Obama Democratic 07/28/2011 Kaplan Lewis A. 03/22/2012 03/23/2012 NaN NaN NaN NaN NaN
1 1 U. S. District Court, Eastern District of New ... USDC Franklin D. Roosevelt Democratic 02/03/1936 new NaN 02/12/1936 02/15/1936 NaN NaN 02/15/1966 05/28/1971 Death
2 2 U. S. District Court, Western District of Penn... USDC Rutherford B. Hayes Republican 01/06/1880 Ketcham Winthrop 01/14/1880 01/14/1880 NaN NaN NaN 02/09/1891 Appointment to Another Judicial Position
3 3 U. S. District Court, Northern District of Ala... USDC Ronald Reagan Republican 07/22/1982 McFadden Frank H. 08/18/1982 08/18/1982 NaN NaN 05/31/1996 NaN NaN
4 4 U. S. District Court, District of New Jersey USDC Jimmy Carter Democratic 09/28/1979 Barlow George H. 10/31/1979 11/02/1979 NaN NaN 02/15/1994 12/02/2009 Death
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
4197 2201 U. S. District Courts, Albemarle, Cape Fear & ... USDC Reassignment Reassignment NaN new NaN NaN 02/13/1801 NaN NaN NaN 03/04/1802 Death
4198 2689 U. S. District Court, Eastern District of Miss... USDC Harry S Truman Democratic 01/13/1949 Harper Roy Winfield 01/31/1949 02/02/1949 NaN NaN 01/05/1971 02/13/1994 Death
4199 1126 U. S. Court of Appeals for the Ninth Circuit USCA William H. Taft Republican 12/12/1910 new NaN 01/31/1911 02/08/1911 NaN NaN 01/31/1928 11/30/1928 Retirement
4200 1453 U. S. Court of Appeals for the Second Circuit USCA Reassignment Reassignment NaN new NaN NaN 07/01/1929 NaN NaN 09/06/1940 09/05/1943 Death
4201 2689 U. S. District Court, Western District of Miss... USDC Harry S Truman Democratic 01/13/1949 Harper Roy Winfield 01/31/1949 02/02/1949 NaN NaN 01/05/1971 02/13/1994 Death

4202 rows × 15 columns

judges_appointments['commission_date'] = pd.to_datetime(judges_appointments['commission_date'])
judges_appointments['commission_date'].dt.year.min()
1789.0
judges_people
judge_id name_first name_middle name_last name_suffix birth_date birthplace_city birthplace_state death_date death_city death_state gender race
0 3419 Ronnie NaN Abrams NaN 1968.0 New York NY NaN NaN NaN F White
1 1 Matthew T. Abruzzo NaN 1889.0 Brooklyn NY 1971.0 Potomac MD M White
2 2 Marcus Wilson Acheson NaN 1828.0 Washington PA 1906.0 Pittsburgh PA M White
3 3 William Marsh Acker Jr. 1927.0 Birmingham AL NaN NaN NaN M White
4 4 Harold Arnold Ackerman NaN 1928.0 Newark NJ 2009.0 West Orange NJ M White
... ... ... ... ... ... ... ... ... ... ... ... ... ...
3527 3391 Jennifer Guerin Zipps NaN 1964.0 Ashland OH NaN NaN NaN F White
3528 2687 Alfonso Joseph Zirpoli NaN 1905.0 Denver CO 1995.0 San Francisco CA M White
3529 2688 William J. Zloch NaN 1944.0 Fort Lauderdale FL NaN NaN NaN M White
3530 2690 Rya Weickert Zobel NaN 1931.0 Zwickau Germany NaN NaN NaN F White
3531 3106 Jack NaN Zouhary NaN 1951.0 Toledo OH NaN NaN NaN M White

3532 rows × 13 columns

judges_people.groupby(["gender"]).size()
gender
F     389
M    3143
dtype: int64
judges_people['race']=judges_people['race'].fillna("Others")
df_grp = judges_people.groupby(["gender","race"]).count().sort_values(["gender","judge_id"], ascending=[True,False])
df_grp
judge_id name_first name_middle name_last name_suffix birth_date birthplace_city birthplace_state death_date death_city death_state
gender race
F White 293 293 265 293 0 293 293 293 28 24 24
African American 51 51 50 51 0 51 51 51 3 3 3
Hispanic 29 29 21 29 0 29 29 29 1 1 1
Asian American 9 9 9 9 0 9 9 9 0 0 0
Others 3 3 3 3 0 3 3 3 0 0 0
African Am./Hispanic 1 1 1 1 0 1 1 1 0 0 0
American Indian 1 1 1 1 0 1 1 1 0 0 0
Hispanic/Asian Am. 1 1 0 1 0 1 1 1 0 0 0
White/Asian Am. 1 1 0 1 0 1 1 1 0 0 0
M White 2871 2871 2483 2871 299 2870 2848 2869 1906 1350 1352
African American 150 150 135 150 41 150 150 150 42 38 38
Hispanic 87 87 70 87 5 87 86 86 15 9 9
Asian American 22 22 18 22 0 22 22 22 6 4 4
Others 6 6 6 6 1 6 6 6 1 1 1
American Indian 2 2 2 2 0 2 2 2 0 0 0
Pac. Isl./Asian Am. 2 2 2 2 0 2 2 2 0 0 0
African Am./Hispanic 1 1 1 1 0 1 1 1 0 0 0
Hispanic/White 1 1 1 1 0 1 1 1 0 0 0
Pac. Isl./White 1 1 1 1 0 1 1 1 1 1 1 
df1 = df_grp["judge_id"].reset_index()
df1
gender race judge_id
0 F White 293
1 F African American 51
2 F Hispanic 29
3 F Asian American 9
4 F Others 3
5 F African Am./Hispanic 1
6 F American Indian 1
7 F Hispanic/Asian Am. 1
8 F White/Asian Am. 1
9 M White 2871
10 M African American 150
11 M Hispanic 87
12 M Asian American 22
13 M Others 6
14 M American Indian 2
15 M Pac. Isl./Asian Am. 2
16 M African Am./Hispanic 1
17 M Hispanic/White 1
18 M Pac. Isl./White 1 
def process_group(group):
    top_rows = group.nlargest(4, 'judge_id')  
    remaining_rows = group.iloc[4:]  
    if not remaining_rows.empty:
        other_sum = remaining_rows['judge_id'].sum()
        other_row = pd.DataFrame({'gender': [group.name], 'race': ['Others'], 'judge_id': [other_sum]})
        return pd.concat([top_rows, other_row], ignore_index=True)
    return top_rows

# Apply function to each group
df_grouped = df1.groupby('gender', group_keys=False).apply(process_group)

df_grouped
gender race judge_id
0 F White 293
1 F African American 51
2 F Hispanic 29
3 F Asian American 9
4 F Others 7
0 M White 2871
1 M African American 150
2 M Hispanic 87
3 M Asian American 22
4 M Others 13 
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.offsetbox import OffsetImage, AnnotationBbox
from PIL import Image
import textwrap

# Define race colors
race_colors = {
    'White': (160, 200, 220), 
    'African American': (70, 130, 180),
    'Hispanic': (0, 0, 128), 
    'Asian American': (30, 144, 255),  
    'Others': (0, 0, 0)  
}
df_grouped_dict = df_grouped.groupby("gender").apply(lambda x: dict(zip(x["race"], x["judge_id"]))).to_dict()

df_percentages = {gender: {race: (count / sum(race_counts.values())) * 100 for race, count in race_counts.items()}
                  for gender, race_counts in df_grouped_dict.items()}

# Function to recolor an icon with stacked race colors
def recolor_icon_layers(icon_path, race_counts, race_colors):
    img = Image.open(icon_path).convert("RGBA")  # Convert to RGBA
    data = np.array(img)  # Convert image to array

    total_count = sum(race_counts.values())
    height = data.shape[0]
    y_start = 0

    for race, count in race_counts.items():
        layer_height = int((count / total_count) * height)
        y_end = y_start + layer_height
        mask = data[y_start:y_end, :, 3] > 0  # Keep only non-transparent pixels
        data[y_start:y_end, :, :-1][mask] = race_colors.get(race, (128, 128, 128))  # Apply race color
        y_start = y_end

    return Image.fromarray(data)

fig, ax = plt.subplots(figsize=(6, 4))

# Calculate total counts for scaling
total_category1 = sum(df_grouped_dict.get("F", {}).values())
total_category2 = sum(df_grouped_dict.get("M", {}).values())

race_colors_mpl = {race: (r/255, g/255, b/255) for race, (r, g, b) in race_colors.items()}

for i, (category, race_counts) in enumerate(df_grouped_dict.items()):
    icon_path = "person-dress.png" if category == "F" else "person.png"
    icon_array = recolor_icon_layers(icon_path, race_counts, race_colors)

    scale_factor = sum(race_counts.values()) / max(total_category1, total_category2)
    base_zoom = 0.8  # Adjust zoom level
    imagebox = OffsetImage(np.array(icon_array), zoom=base_zoom * scale_factor)

    ab = AnnotationBbox(imagebox, (i, 0), frameon=False, xycoords="data", box_alignment=(0.5, 0))
    ax.add_artist(ab)

    percentages = df_percentages[category]
    for j, (race, percent) in enumerate(reversed(percentages.items())):
        if(category == "F"):
            ax.text(i+0.3, j*150, f"{percent:.1f}%", ha='right', fontsize=10, color=race_colors_mpl[race])
        else:            
            ax.text(i+0.4, j*250, f"{race}: {percent:.1f}%", ha='left', fontsize=10, color=race_colors_mpl[race])

ax.text(0, 1000, f"Female judges: {total_category1}", ha='center', fontsize=10, color="black")
ax.text(1, 4000, f"Male judges: {total_category2:,}", ha='center', fontsize=10, color="black")

ax.set_xticks([0, 1])
ax.set_xlim(-0.5, 1.5)
ax.set_ylim(0, max(total_category1, total_category2) + 100)
title = f"Among the US judges appointed from {judges_appointments['commission_date'].dt.year.min():.0f} to {judges_appointments['commission_date'].dt.year.max():.0f}, there were about 8 male judges for every female judge. The proportion of the top four races is shown in different colors."
wrapped_title = "\n".join(textwrap.wrap(title, width=30))
ax.set_title(wrapped_title, loc='left', pad=75)
ax.axis("off")
fig.patch.set_facecolor("whitesmoke")
fig.savefig("judges_appoint.png", bbox_inches="tight", dpi=300)
plt.show()