
!git clone https://github.com/eugenelim17/eugenelim17.github.io

Cloning into 'eugenelim17.github.io'...
remote: Enumerating objects: 133, done.
remote: Counting objects: 100% (50/50), done.
remote: Compressing objects: 100% (47/47), done.
remote: Total 133 (delta 28), reused 3 (delta 3), pack-reused 83
Receiving objects: 100% (133/133), 23.85 MiB | 25.36 MiB/s, done.
Resolving deltas: 100% (65/65), done.


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

# Displays a greater number of columns when dataframes are displayed
pd.set_option("display.max_columns", None)

# Reads in the first dataframe
job_postings_df = pd.read_csv("eugenelim17.github.io/job_postings.csv.zip", compression="zip")

# First dataframe before any cleaning or modification
job_postings_df


# Reads in the second dataframe
salaries_df = pd.read_csv("eugenelim17.github.io/2302023 Raw Data - Ask A Manager Salary Survey 2021.csv")

# Second dataframe before any cleaning or modification
salaries_df


# Displays first dataframe again
job_postings_df


# Dataframe data types before any adjustment to the types
job_postings_df.dtypes

job_id                          int64
company_id                    float64
title                          object
description                    object
max_salary                    float64
med_salary                    float64
min_salary                    float64
pay_period                     object
formatted_work_type            object
location                       object
applies                       float64
original_listed_time          float64
remote_allowed                float64
views                         float64
job_posting_url                object
application_url                object
application_type               object
expiry                        float64
closed_time                   float64
formatted_experience_level     object
skills_desc                    object
listed_time                   float64
posting_domain                 object
sponsored                       int64
work_type                      object
currency                       object
compensation_type              object
dtype: object


job_postings_df["job_id"] = job_postings_df["job_id"].astype(str)
job_postings_df["company_id"] = job_postings_df["company_id"].astype(str)


# Dataframe data types after adjustment to some of the types
job_postings_df.dtypes

job_id                         object
company_id                     object
title                          object
description                    object
max_salary                    float64
med_salary                    float64
min_salary                    float64
pay_period                     object
formatted_work_type            object
location                       object
applies                       float64
original_listed_time          float64
remote_allowed                float64
views                         float64
job_posting_url                object
application_url                object
application_type               object
expiry                        float64
closed_time                   float64
formatted_experience_level     object
skills_desc                    object
listed_time                   float64
posting_domain                 object
sponsored                       int64
work_type                      object
currency                       object
compensation_type              object
dtype: object


# Confirms that all observations without a compensation type value also have no maximum, median, or minimum salary
job_postings_df[job_postings_df['compensation_type'].isnull()]
job_postings_df[job_postings_df['compensation_type'].isnull()]["max_salary"].isnull().sum()
job_postings_df[job_postings_df['compensation_type'].isnull()]["med_salary"].isnull().sum()
job_postings_df[job_postings_df['compensation_type'].isnull()]["min_salary"].isnull().sum()

# Rows that do not have a compensation type are removed here using .notna()
job_postings_df = job_postings_df[job_postings_df['compensation_type'].notna()].copy()


# Any salary that is on an hourly basis is adjusted to a yearly salary by multiplying by 40 since that is the American national standard for weekly hours worked
# and then multiplying by 52 because there are 52 weeks in a year.
job_postings_df['max_salary'] = job_postings_df.apply(lambda row: row['max_salary'] * 40 * 52 if row['pay_period'] == 'HOURLY' else row['max_salary'], axis=1)
job_postings_df['med_salary'] = job_postings_df.apply(lambda row: row['med_salary'] * 40 * 52 if row['pay_period'] == 'HOURLY' else row['med_salary'], axis=1)
job_postings_df['min_salary'] = job_postings_df.apply(lambda row: row['min_salary'] * 40 * 52 if row['pay_period'] == 'HOURLY' else row['min_salary'], axis=1)

job_postings_df['max_salary'] = job_postings_df.apply(lambda row: row['max_salary'] * 12 if row['pay_period'] == 'MONTHLY' else row['max_salary'], axis=1)
job_postings_df['med_salary'] = job_postings_df.apply(lambda row: row['med_salary'] * 12 if row['pay_period'] == 'MONTHLY' else row['med_salary'], axis=1)
job_postings_df['min_salary'] = job_postings_df.apply(lambda row: row['min_salary'] * 12 if row['pay_period'] == 'MONTHLY' else row['min_salary'], axis=1)

job_postings_df['pay_period'] = 'Yearly'


# Fill the "remote_allowed" column for in-person workers with a value of 0
job_postings_df['remote_allowed'] = job_postings_df['remote_allowed'].fillna(0)


job_postings_df.duplicated().sum()

0


# Displays first dataframe after cleaning
job_postings_df


# Makes average salary column because controlled analysis has too few observations for "med_salary"
job_postings_df["avg_salary"] = (job_postings_df["max_salary"] + job_postings_df["min_salary"]) / 2

# Create new filtered dataframes to help calculate summary statistics for remote and in-person workers based on yearly median salary
remote_workers = job_postings_df[job_postings_df['remote_allowed'] == 1]
in_person = job_postings_df[job_postings_df['remote_allowed'] == 0]

# Summary statistics for remote workers
summary_remote = remote_workers['avg_salary'].describe()

# Summary statistics for in-person workers
summary_in_person = in_person['avg_salary'].describe()

# Print the summary statistics
print("Summary Statistics for Remote Positions:")
print(summary_remote)

# Setting option to display numbers in normal notation
pd.set_option('display.float_format', lambda x: '%.3f' % x)

print("\nSummary Statistics for In-Person Positions:")
print(summary_in_person)

Summary Statistics for Remote Positions:
count       897.000000
mean     115667.544621
std       65507.056828
min          23.000000
25%       75000.000000
50%      109200.000000
75%      145600.000000
max      875000.000000
Name: avg_salary, dtype: float64

Summary Statistics for In-Person Positions:
count      4624.000
mean      95899.397
std       56219.278
min          13.500
25%       55000.000
50%       82500.000
75%      122790.000
max     1050000.000
Name: avg_salary, dtype: float64


# Assuming 'remote_allowed' is the column indicating remote work and 'title' is the column containing job titles
# Filter data for remote and in-person workers
remote_workers = job_postings_df[job_postings_df['remote_allowed'] == 1]
in_person = job_postings_df[job_postings_df['remote_allowed'] == 0]

# Get unique job titles for remote and in-person workers
remote_job_titles = set(remote_workers['title'].unique())
in_person_job_titles = set(in_person['title'].unique())

# Find common job titles between remote and in-person workers
common_job_titles = remote_job_titles.intersection(in_person_job_titles)

# DataFrames for common job titles
remote_common_jobs = remote_workers[remote_workers['title'].isin(common_job_titles)].copy()
in_person_common_jobs = in_person[in_person['title'].isin(common_job_titles)].copy()

print("Common Job Titles:")
print(common_job_titles)

# Print the summary statistics for remote and in-person workers with common job titles
print("\nSummary Statistics for Remote Positions with Common Job Titles:")
print(remote_common_jobs['avg_salary'].describe())

print("\nSummary Statistics for In-Person Positions with Common Job Titles:")
print(in_person_common_jobs['avg_salary'].describe())

Common Job Titles:
{'Human Resources Business Partner', 'Digital Marketing Specialist', 'Video Editor', 'Quality Assurance Analyst', 'Senior Test Engineer', 'Systems Administrator', 'Business Development Representative', 'Senior Financial Analyst', 'Compliance Specialist', 'Scoring Content Specialist', 'Corporate Accountant', 'Sales Account Manager', 'Human Resources Assistant', 'Remote Medical Device Supplier Engineer', 'Senior Accountant', 'Senior Tax Manager', 'Contract Attorney', 'Information Security Analyst', 'Sales And Marketing Specialist', 'Field Operations Manager', 'Logistics Coordinator', 'Senior Software Engineer', 'Production Artist', 'Human Resources Director', 'Senior Business Analyst', 'Technical Writer', 'Network Engineer', 'B2B Sales Representative', 'Mental Health Therapist', 'Analyst, Compensation Operations', 'Senior Data Analyst', 'Senior Data Engineer', 'Full Stack Engineer', 'Director of Product, MyBSWHealth', 'Senior Product Designer', 'Network Administrator', 'Senior Project Manager', 'Senior Revenue Accountant', 'Sales Account Executive', 'Program Manager', 'UI/UX Designer', 'Director of Information Technology', 'Maintenance Technician', 'Customer Service Representative', 'Sales Manager', 'Business Development Executive', 'Director of Development', 'Executive Administrator', 'Product Marketing Manager', 'Attorney', 'Tax Manager', 'Customer Success Manager', 'Legal Assistant Paralegal', 'Program Assistant', 'Bookkeeper', 'Account Executive', 'Accounting Manager', 'HR Generalist', 'Cloud Architect', 'Support Engineer', 'Business Analyst', 'Sales Development Representative', 'Regional Sales Manager', 'Controller', 'Field Service Technician', 'Staff Accountant', 'Account Manager', 'Director of Sales', 'Sales', 'Recruiter', 'Scrum Master', 'Accounts Payable Specialist', 'Design Engineer', 'Sales Representative', 'Senior Systems Engineer', 'Architect', 'Senior Manager, R&D', 'Software Engineer', 'Research Scientist', 'Project Manager', 'Oracle Database Administrator', 'Data Engineer', 'Human Resource Information System Manager', 'Accountant', 'Customer Service Specialist', 'Medical Strategy Senior Manager (MRD Lead)', 'Process Engineer', 'Social Media Manager', 'Sales Associate', 'Financial Controller', 'Inside Sales Representative', 'Power BI Developer', 'Business Intelligence Analyst', 'Business Development Specialist', 'Graphic Designer', 'Global Alliance Director, Salesforce'}

Summary Statistics for Remote Positions with Common Job Titles:
count      153.000
mean    102119.660
std      43441.857
min      24000.000
25%      67500.000
50%      92500.000
75%     143300.000
max     209850.000
Name: avg_salary, dtype: float64

Summary Statistics for In-Person Positions with Common Job Titles:
count      361.000
mean     94210.016
std      44317.269
min         16.000
25%      57500.000
50%      87500.000
75%     119600.000
max     375000.000
Name: avg_salary, dtype: float64


job_postings_df['med_salary'].plot.hist(bins=50, color='blue', alpha=0.7, edgecolor='black', title='Distribution of Median Yearly Salaries')
plt.xlabel('Median Yearly Salary')
plt.ylabel('Frequency')
plt.show()


job_postings_df['med_salary_log'] = np.log1p(job_postings_df['med_salary'])
job_postings_df['med_salary_log'].plot.hist(bins=50, color='blue', alpha=0.7, edgecolor='black', title='Distribution of Median Yearly Salaries (Log-Transformed)')

# Set labels and title
plt.xlabel('Log(Median Yearly Salary)')
plt.ylabel('Frequency')
plt.show()


# Define the mapping from job titles to industries
job_title_to_industry_mapping = {
    "Warehouse Order Selector": "Warehouse",
    "Package Handler (Warehouse like)": "Warehouse",
    "Warehouse Associate": "Warehouse",
    "Warehouse Worker - SAS Safety Corp.": "Warehouse",
    "Equipment Technician": "Technician",
    "Field Service Technician": "Technician",
    "Pest Control Technician": "Technician",
    "Central Services Technician": "Technician"
}

# Use the .replace() method to create the 'industries' column
job_postings_df['industries'] = job_postings_df['title'].replace(job_title_to_industry_mapping)


selected_industries = ["Warehouse", "Technician"]
filtered_df = job_postings_df[job_postings_df['industries'].isin(selected_industries)].copy()

# Summary statistics for "med_salary" for each industry
summary_statistics = filtered_df.groupby('industries')['med_salary'].describe()

# Print the summary statistics
print(summary_statistics)

            count      mean       std       min       25%       50%       75%  \
industries                                                                      
Technician 11.000 50196.218 12715.308 41600.000 42500.000 42640.000 50679.200   
Warehouse  16.000 41233.800 12502.981 31200.000 33280.000 38740.000 41459.600   

                 max  
industries            
Technician 75000.000  
Warehouse  82000.000


# Displays first dataframe after preliminary EDA
job_postings_df


# Displays the second dataset again in its original state
salaries_df


# Changes column titles
salaries_df.columns = ["Timestamp", "Age", "Industry", "Job Title", "Additional Job Title Context", "Annual Salary", "Additional Compensation", "Currency",
                       "Other Currency Type", "Additional Income Context", "Country", "U.S. State", "City", "Professional Experience Overall",
                       "Professional Experience In Field", "Education", "Gender", "Race"]

salaries_df


# Dataframe data types before any adjustment to the types
salaries_df.dtypes

Timestamp                            object
Age                                  object
Industry                             object
Job Title                            object
Additional Job Title Context         object
Annual Salary                        object
Additional Compensation             float64
Currency                             object
Other Currency Type                  object
Additional Income Context            object
Country                              object
U.S. State                           object
City                                 object
Professional Experience Overall      object
Professional Experience In Field     object
Education                            object
Gender                               object
Race                                 object
dtype: object


salaries_df["Annual Salary"] = salaries_df["Annual Salary"].str.replace(",","").astype(float)


# Dataframe data types after adjustment to one of the types
salaries_df.dtypes

Timestamp                            object
Age                                  object
Industry                             object
Job Title                            object
Additional Job Title Context         object
Annual Salary                       float64
Additional Compensation             float64
Currency                             object
Other Currency Type                  object
Additional Income Context            object
Country                              object
U.S. State                           object
City                                 object
Professional Experience Overall      object
Professional Experience In Field     object
Education                            object
Gender                               object
Race                                 object
dtype: object


# Removes salaries not in USD
salaries_df = salaries_df[salaries_df["Currency"] == "USD"].copy()


# Checks which columns have null values
salaries_df.isnull().sum()

Timestamp                               0
Age                                     0
Industry                               57
Job Title                               0
Additional Job Title Context        17172
Annual Salary                           0
Additional Compensation              5826
Currency                                0
Other Currency Type                 23255
Additional Income Context           20751
Country                                 0
U.S. State                            350
City                                   59
Professional Experience Overall         0
Professional Experience In Field        0
Education                             151
Gender                                138
Race                                  143
dtype: int64


salaries_df.dropna(subset=["U.S. State", "City", "Education", "Gender", "Race"], inplace=True)
salaries_df.reset_index(drop=True, inplace=True)


# Gets total number of duplicated observations
salaries_df.duplicated().sum()

0


# Displays the second dataframe after cleaning
salaries_df


male_salaries = salaries_df[salaries_df['Gender'] == 'Man']['Annual Salary']
female_salaries = salaries_df[salaries_df['Gender'] == 'Woman']['Annual Salary']

male_salary_stats = male_salaries.describe()
female_salary_stats = female_salaries.describe()

# Setting options to display numbers in normal notation
pd.set_option('display.float_format', lambda x: '%.3f' % x)

# Printing the results
print("\nMale Salary Statistics:")
print(male_salary_stats)

print("\nFemale Salary Statistics:")
print(female_salary_stats)

Male Salary Statistics:
count      4147.000
mean     114475.172
std       76252.431
min           0.000
25%       69000.000
50%      100040.000
75%      148000.000
max     1650000.000
Name: Annual Salary, dtype: float64

Female Salary Statistics:
count     17596.000
mean      87049.734
std       69253.195
min           0.000
25%       55000.000
50%       75000.000
75%      104217.250
max     5000044.000
Name: Annual Salary, dtype: float64


pd.crosstab(salaries_df["Gender"], salaries_df["Education"], margins=True)


pd.crosstab(salaries_df["Gender"], salaries_df["Education"], normalize=True, margins=True)


joint = pd.crosstab(salaries_df["Gender"], salaries_df["Education"], normalize=True)

sns.heatmap(joint)

<Axes: xlabel='Education', ylabel='Gender'>


# Displays first dataframe after preliminary EDA
salaries_df


# Count the number of occurrences of each unique job title in the 'title' column, select the highest ten and create a bar graph
job_postings_df["title"].value_counts()[[1,2,3,4,6,7,8,9,10,12]].plot.bar(ylabel="Count", xlabel="Job Title", title="Count by Job Title")

<Axes: title={'center': 'Count by Job Title'}, xlabel='Job Title', ylabel='Count'>


filter2 = ((job_postings_df["title"] == "Account Executive") | (job_postings_df["title"] == "Retail Sales Associate") | (job_postings_df["title"] == "Call Center Support Rep") |
          (job_postings_df["title"] == "Scoring Content Specialist") | (job_postings_df["title"] == "Administrative Assistant") | (job_postings_df["title"] == "Sales Manager") |
          (job_postings_df["title"] == "Staff Accountant") | (job_postings_df["title"] == "Senior Accountant") | (job_postings_df["title"] == "Controller") |
          (job_postings_df["title"] == "Executive Assistant"))

job_postings_df[filter2].groupby("title")["avg_salary"].mean().sort_values(ascending=False).plot.bar(ylabel="Average Salary", xlabel="Job Title", title="Average Salary by Job Title")

<Axes: title={'center': 'Average Salary by Job Title'}, xlabel='Job Title', ylabel='Average Salary'>


# Count the number of occurrences of each unique location in the 'location' column, select the highest ten and create a bar graph
job_postings_df["location"].value_counts()[[1,2,3,4,5,7,8,9,10,11]].plot.bar(ylabel="Count", xlabel="City", title="Count by U.S. City")

<Axes: title={'center': 'Count by U.S. City'}, xlabel='City', ylabel='Count'>


# Create a boolean filter for specific locations
filter1 = ((job_postings_df["location"] == "New York, NY") | (job_postings_df["location"] == "Los Angeles, CA") | (job_postings_df["location"] == "Seattle, WA") |
          (job_postings_df["location"] == "San Diego, CA") | (job_postings_df["location"] == "San Francisco, CA") | (job_postings_df["location"] == "Denver, CO") |
          (job_postings_df["location"] == "Chicago, IL") | (job_postings_df["location"] == "Atlanta, GA") | (job_postings_df["location"] == "Houston, TX") |
          (job_postings_df["location"] == "Dallas, TX"))

# Filter the DataFrame based on the specified locations and calculate the mean of 'avg_salary' for each location, then sort the average salaries in descending order and plot a bar chart
job_postings_df[filter1].groupby("location")["avg_salary"].mean().sort_values(ascending=False).plot.bar(ylabel="Average Salary", xlabel="City", title="Average Salary by U.S. City")

<Axes: title={'center': 'Average Salary by U.S. City'}, xlabel='City', ylabel='Average Salary'>


# Extract the state code from the 'location' column and create a new 'state' column
job_postings_df["state"] = job_postings_df["location"].str[-2:]

# Keep only the valid state codes in the 'state' column
job_postings_df["state"] = job_postings_df["state"].where(job_postings_df["state"].isin(["AL", "AK", "AZ", "AR", "CA", "CO", "CT", "DE", "DC", "FL", "GA", "HI",
                                                                                         "ID", "IL", "IN", "IA", "KS", "KY", "LA", "ME", "MD", "MA", "MI", "MN", "MS", "MO", "MT", "NE", "NV", "NH", "NJ", "NM", "NY", "NC", "ND", "OH", "OK", "OR", "PA", "RI", "SC", "SD", "TN", "TX", "UT", "VT", "VA", "VI", "WA", "WV", "WI", "WY"]))
# Group by state and calculate the mean of 'avg_salary' for each state, then plot horizontal bar chart
job_postings_df.groupby("state")["avg_salary"].mean().sort_values(ascending=False).plot.barh(ylabel="Average Salary", xlabel="State", figsize=(8,9), title="Average Salary by U.S. State")

<Axes: title={'center': 'Average Salary by U.S. State'}, xlabel='State', ylabel='Average Salary'>


state_code = 'LA'
louisiana_data = job_postings_df[job_postings_df["state"] == state_code]
print("Summary statistics for Louisiana:")
print(louisiana_data["avg_salary"].describe())

Summary statistics for Louisiana:
count       20.000
mean    115095.500
std     111072.655
min      22880.000
25%      48882.000
50%      85870.000
75%     108125.000
max     362500.000
Name: avg_salary, dtype: float64


state_code = 'LA'
louisiana_data = job_postings_df[job_postings_df["state"] == state_code]
sorted_louisiana_data = louisiana_data.sort_values(by='avg_salary', ascending=False)
highest_salaries_in_louisiana = sorted_louisiana_data[['title', 'avg_salary']].head(10).values.tolist()
for title, salary in highest_salaries_in_louisiana:
    print(f"{title}: ${salary:,.2f}")

Branch Leader: $362,500.00
Branch Leader: $362,500.00
Branch Leader: $362,500.00
Industrial Construction Superintendents: $130,000.00
Project Engineer: $117,500.00
Associate Attorney (New Orleans): $105,000.00
Business Development Manager: $105,000.00
Program Manager 3--DHH(Clinical Systems and Services Team Lead): $104,112.00
Pipeline Construction Supervisor: $90,000.00
Auto Body Technician: $87,500.00


state_code = 'LA'
louisiana_data = job_postings_df[job_postings_df["state"] == state_code]
sorted_louisiana_data = louisiana_data.sort_values(by='avg_salary', ascending=False)
trimmed_louisiana_data = sorted_louisiana_data.iloc[3:]
highest_salaries_in_louisiana = trimmed_louisiana_data[['title', 'avg_salary']].head(10).values.tolist()
for title, salary in highest_salaries_in_louisiana:
    print(f"{title}: ${salary:,.2f}")

Industrial Construction Superintendents: $130,000.00
Project Engineer: $117,500.00
Associate Attorney (New Orleans): $105,000.00
Business Development Manager: $105,000.00
Program Manager 3--DHH(Clinical Systems and Services Team Lead): $104,112.00
Pipeline Construction Supervisor: $90,000.00
Auto Body Technician: $87,500.00
Cleared Senior Information Technology Specialist: $84,240.00
MEDICAID ANALYST SUPERVISOR: $67,476.00
PC Technician: $50,960.00


# Calculate summary statistics for the 'avg_salary' column in the modified Louisiana data
summary_statistics_louisiana = trimmed_louisiana_data["avg_salary"].describe()
print(summary_statistics_louisiana)

count       17.000
mean     71435.882
std      33887.537
min      22880.000
25%      47136.000
50%      67476.000
75%     104112.000
max     130000.000
Name: avg_salary, dtype: float64


state_code = 'LA'

# Create a copy of the original DataFrame
job_postings_copy = job_postings_df.copy()

# Filter the DataFrame for records related to Louisiana
louisiana_data = job_postings_copy[job_postings_copy["state"] == state_code]

# Sort the Louisiana data by 'avg_salary' in descending order
sorted_louisiana_data = louisiana_data.sort_values(by='avg_salary', ascending=False)

# Remove the top 3 entries
trimmed_louisiana_data = sorted_louisiana_data.iloc[3:]

# Update the salaries for Louisiana with the trimmed data
job_postings_copy.loc[job_postings_copy["state"] == state_code, "avg_salary"] = trimmed_louisiana_data["avg_salary"]

# Group by state and calculate the mean of 'avg_salary' for each state
average_salary_by_state = job_postings_copy.groupby("state")["avg_salary"].mean()

# Sort the average salaries in descending order and plot horizontal bar chart
average_salary_by_state.sort_values(ascending=False).plot.barh(ylabel="Average Salary", xlabel="State", figsize=(8, 9), title="Average Salary by U.S. State (Louisiana - Top 3 Removed)")

<Axes: title={'center': 'Average Salary by U.S. State (Louisiana - Top 3 Removed)'}, xlabel='State', ylabel='Average Salary'>


import seaborn as sns
import matplotlib.pyplot as plt

# Extract the state code from the 'location' column and create a new 'state' column
job_postings_df["state"] = job_postings_df["location"].str[-2:].copy()

# Define a dictionary mapping each state to its region
state_to_region = {
    "AL": "South", "AK": "West", "AZ": "West", "AR": "South", "CA": "West", "CO": "West", "CT": "Northeast",
    "DE": "South", "DC": "South", "FL": "South", "GA": "South", "HI": "West", "ID": "West", "IL": "Midwest",
    "IN": "Midwest", "IA": "Midwest", "KS": "Midwest", "KY": "South", "LA": "South", "ME": "Northeast",
    "MD": "South", "MA": "Northeast", "MI": "Midwest", "MN": "Midwest", "MS": "South", "MO": "Midwest",
    "MT": "West", "NE": "Midwest", "NV": "West", "NH": "Northeast", "NJ": "Northeast", "NM": "West",
    "NY": "Northeast", "NC": "South", "ND": "Midwest", "OH": "Midwest", "OK": "South", "OR": "West",
    "PA": "Northeast", "RI": "Northeast", "SC": "South", "SD": "Midwest", "TN": "South", "TX": "South",
    "UT": "West", "VT": "Northeast", "VA": "South", "VI": "South", "WA": "West", "WV": "South",
    "WI": "Midwest", "WY": "West"
}

# Create a new 'region' column based on the mapping
job_postings_df["region"] = job_postings_df["state"].map(state_to_region).copy()

# Keep only the valid state codes in the 'state' column
job_postings_df = job_postings_df[job_postings_df["state"].isin(state_to_region.keys())].copy()

# Exclude the top three salaries for Louisiana
louisiana_salaries = job_postings_df[job_postings_df["state"] == "LA"]["avg_salary"]
top_louisiana_salaries = louisiana_salaries.nlargest(3)
job_postings_df.loc[(job_postings_df["state"] == "LA") & (job_postings_df["avg_salary"].isin(top_louisiana_salaries.index)), "avg_salary"] = louisiana_salaries[~louisiana_salaries.isin(top_louisiana_salaries)].mean()

# Group by region and calculate the mean of 'avg_salary' for each region
region_avg_salary = job_postings_df.groupby("region")["avg_salary"].mean().sort_values(ascending=False)

# Define a color palette for each region
region_colors = {"West": "blue", "Northeast": "green", "Midwest": "orange", "South": "red"}

sns.barplot(x=region_avg_salary, y=region_avg_salary.index, palette=region_colors)

plt.xlabel("Average Salary")
plt.ylabel("Region")
plt.title("Average Salary by Region (Top 3 Salaries of Louisiana Removed)")
plt.show()


job_postings_df["formatted_experience_level"].value_counts().plot.bar(ylabel="Count", xlabel="Work Experience Level", title="Count by Work Experience Level")

<Axes: title={'center': 'Count by Work Experience Level'}, xlabel='Work Experience Level', ylabel='Count'>


job_postings_df.groupby("formatted_experience_level")["avg_salary"].mean().sort_values(ascending=False).plot.bar(ylabel="Average Salary", xlabel="Work Experience Level", title="Average Salary by Work Expereince Level")

<Axes: title={'center': 'Average Salary by Work Expereince Level'}, xlabel='Work Experience Level', ylabel='Average Salary'>


job_postings_df["formatted_experience_level"].value_counts()

Mid-Senior level    1678
Entry level         1229
Associate            477
Director             240
Executive             51
Internship            46
Name: formatted_experience_level, dtype: int64


salaries_df["Professional Experience In Field"].value_counts()

11 - 20 years       5329
5-7 years           5283
2 - 4 years         4860
8 - 10 years        4071
21 - 30 years       1534
1 year or less      1119
31 - 40 years        324
41 years or more      32
Name: Professional Experience In Field, dtype: int64


salaries_df["Mapped Experience (In Field)"] = salaries_df["Professional Experience In Field"].map({"1 year or less": "Internship", "2 - 4 years": "Entry level", "5-7 years": "Associate", "8 - 10 years": "Mid-Senior level", "11 - 20 years": "Director", "21 - 30 years": "Director", "31 - 40 years": "Executive", "41 years or more": "Executive"}).copy()


import pandas as pd

# Create a dictionary to map full state names to abbreviations
state_abbr = {
    'Alabama': 'AL', 'Alaska': 'AK', 'Arizona': 'AZ', 'Arkansas': 'AR', 'California': 'CA',
    'Colorado': 'CO', 'Connecticut': 'CT', 'District of Columbia': 'DC', 'Delaware': 'DE', 'Florida': 'FL', 'Georgia': 'GA',
    'Hawaii': 'HI', 'Idaho': 'ID', 'Illinois': 'IL', 'Indiana': 'IN', 'Iowa': 'IA', 'Kansas': 'KS',
    'Kentucky': 'KY', 'Louisiana': 'LA', 'Maine': 'ME', 'Maryland': 'MD', 'Massachusetts': 'MA',
    'Michigan': 'MI', 'Minnesota': 'MN', 'Mississippi': 'MS', 'Missouri': 'MO', 'Montana': 'MT',
    'Nebraska': 'NE', 'Nevada': 'NV', 'New Hampshire': 'NH', 'New Jersey': 'NJ', 'New Mexico': 'NM',
    'New York': 'NY', 'North Carolina': 'NC', 'North Dakota': 'ND', 'Ohio': 'OH', 'Oklahoma': 'OK',
    'Oregon': 'OR', 'Pennsylvania': 'PA', 'Rhode Island': 'RI', 'South Carolina': 'SC', 'South Dakota': 'SD',
    'Tennessee': 'TN', 'Texas': 'TX', 'Utah': 'UT', 'Vermont': 'VT', 'Virginia': 'VA', 'Washington': 'WA',
    'West Virginia': 'WV', 'Wisconsin': 'WI', 'Wyoming': 'WY'
}

# Create a new column for state abbreviations
salaries_df['State Abbreviation'] = salaries_df['U.S. State'].map(state_abbr)

# Create a new column for location in the format 'city, state abbreviation'
salaries_df['Location'] = salaries_df['City'] + ', ' + salaries_df['State Abbreviation']

salaries_df


job_postings_df.isnull().sum()

job_id                           0
company_id                       0
title                            0
description                      0
max_salary                     887
med_salary                    4469
min_salary                     887
pay_period                       0
formatted_work_type              0
location                         0
applies                       2254
original_listed_time             0
remote_allowed                   0
views                          834
job_posting_url                  0
application_url               1974
application_type                 0
expiry                           0
closed_time                   5141
formatted_experience_level    1635
skills_desc                   5299
listed_time                      0
posting_domain                2242
sponsored                        0
work_type                        0
currency                         0
compensation_type                0
avg_salary                     887
med_salary_log                4469
industries                       0
state                            0
region                           0
dtype: int64


salaries_df.isnull().sum()

Timestamp                               0
Age                                     0
Industry                               55
Job Title                               0
Additional Job Title Context        16603
Annual Salary                           0
Additional Compensation              5578
Currency                                0
Other Currency Type                 22528
Additional Income Context           20089
Country                                 0
U.S. State                              0
City                                    0
Professional Experience Overall         0
Professional Experience In Field        0
Education                               0
Gender                                  0
Race                                    0
Mapped Experience (In Field)            0
State Abbreviation                    105
Location                              105
dtype: int64


# Drop rows with NaNs in "formatted_experience_level" and "avg_salary"
job_postings_df.dropna(subset=["formatted_experience_level", "avg_salary"], inplace=True)
salaries_df.dropna(subset=["Location"], inplace=True)


# Makes all the characters in the job titles of both datasets lowercase
job_postings_df['title'] = job_postings_df['title'].str.lower()
salaries_df['Job Title'] = salaries_df['Job Title'].str.lower()


job_postings_df.reset_index(inplace=True)


necessary_columns = job_postings_df[["title", "avg_salary", "location", "formatted_experience_level"]]
necessary_columns


titles = list(necessary_columns["title"])


from sklearn.feature_extraction.text import CountVectorizer

vectorizer = CountVectorizer(binary=True)
vectorizer.fit(titles)

title_words_binary = pd.DataFrame(vectorizer.transform(titles).toarray())
title_words_binary.columns = vectorizer.get_feature_names_out()
title_words_binary


model_input = pd.concat([necessary_columns, title_words_binary], axis=1)
model_input


salaries_df["location"] = salaries_df["Location"].copy()
salaries_df["formatted_experience_level"] = salaries_df["Mapped Experience (In Field)"].copy()
salaries_df["title"] = salaries_df["Job Title"].copy()


salaries_df.reset_index(inplace=True)


necessary_columns_2 = salaries_df[["title", "location", "formatted_experience_level"]]
necessary_columns_2


titles_2 = list(necessary_columns_2["title"])


vectorizer = CountVectorizer(binary=True)
vectorizer.fit(titles_2)

title_words_binary_2 = pd.DataFrame(vectorizer.transform(titles_2).toarray())
title_words_binary_2.columns = vectorizer.get_feature_names_out()
title_words_binary_2


model_input_2 = pd.concat([necessary_columns_2, title_words_binary_2], axis=1)
model_input_2


from sklearn.feature_extraction import DictVectorizer
from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import KNeighborsRegressor

X_train_dict = model_input.iloc[:, 2:].to_dict(orient="records")
X_test_dict = model_input_2.iloc[:, 1:].to_dict(orient="records")
y_train = model_input["avg_salary"]

# Dummy encoding
vec = DictVectorizer(sparse=False)
vec.fit(X_train_dict)
X_train = vec.transform(X_train_dict)
X_test = vec.transform(X_test_dict)

# Standardization
scaler = StandardScaler()
scaler.fit(X_train)
X_train_sc = scaler.transform(X_train)
X_test_sc = scaler.transform(X_test)

# K-Nearest Neighbors Model
model = KNeighborsRegressor(n_neighbors=10)
model.fit(X_train_sc, y_train)
y_pred = model.predict(X_test_sc)

<ipython-input-67-bfea8513a9f7>:6: UserWarning: DataFrame columns are not unique, some columns will be omitted.
  X_test_dict = model_input_2.iloc[:, 1:].to_dict(orient="records")


pd.DataFrame(y_pred)


salaries_df.shape

(22447, 25)


salaries_df["Predicted Salary"] = y_pred
salaries_df["Salary Delta"] = (salaries_df["Annual Salary"] - salaries_df["Predicted Salary"]).abs()
salaries_df["Salary Delta"].mean()

35657.02563763533


def get_salary_delta_avg(k):
    model = KNeighborsRegressor(n_neighbors=k)
    model.fit(X_train_sc, y_train)
    y_pred = model.predict(X_test_sc)

    salaries_df["Predicted Salary (Hyper)"] = y_pred
    salaries_df["Salary Delta (Hyper)"] = (salaries_df["Annual Salary"] - salaries_df["Predicted Salary (Hyper)"]).abs()
    result = salaries_df["Salary Delta (Hyper)"].mean()
    return result

ks = pd.Series(range(1, 31))
ks.index = range(1, 31)
test_errs = ks.apply(get_salary_delta_avg)

test_errs.plot.line()
test_errs.sort_values()

10   35657.026
9    35843.094
11   35845.345
12   35927.718
13   35959.211
8    36131.327
14   36164.220
15   36310.537
7    36716.141
16   36759.526
17   37440.884
6    37729.531
18   37995.385
19   38449.488
20   38848.708
5    39029.222
21   39290.687
22   39522.937
23   39975.685
24   40156.661
25   40405.044
26   40552.819
27   40688.536
4    40818.538
28   40820.853
29   40838.966
30   40954.979
3    42621.922
2    43687.182
1    49780.924
dtype: float64


salaries_df.groupby("Gender")["Salary Delta"].mean().sort_values(ascending=False).plot.bar(ylabel="Salary Delta", xlabel="Gender", title="Difference Between Predicted and Actual Salary by Gender")

<Axes: title={'center': 'Difference Between Predicted and Actual Salary by Gender'}, xlabel='Gender', ylabel='Salary Delta'>


salaries_df.groupby("Gender")["Salary Delta"].mean()[[0,1,4]].mean()

37889.78239898508


salaries_df.groupby("Age")["Salary Delta"].mean().sort_values(ascending=False).plot.bar(ylabel="Salary Delta", xlabel="Age", title="Difference Between Predicted and Actual Salary by Age")

<Axes: title={'center': 'Difference Between Predicted and Actual Salary by Age'}, xlabel='Age', ylabel='Salary Delta'>


salaries_df.groupby("Age")["Salary Delta"].mean().mean()

39252.9279575972


salaries_df.groupby("Education")["Salary Delta"].mean().sort_values(ascending=False).plot.bar(ylabel="Salary Delta", xlabel="Education", title="Difference Between Predicted and Actual Salary by Education")

<Axes: title={'center': 'Difference Between Predicted and Actual Salary by Education'}, xlabel='Education', ylabel='Salary Delta'>


salaries_df.groupby("Education")["Salary Delta"].mean().mean()

40769.758876291795


selected_columns = salaries_df[["Job Title", "Location", "Mapped Experience (In Field)", "Age", "Gender", "Education", "Predicted Salary", "Annual Salary", "Salary Delta"]]
selected_columns.sort_values(by=["Salary Delta"])


all_answered_norm = salaries_df[~((salaries_df["Gender"] == "Prefer not to answer") | (salaries_df["Gender"] == "Other or prefer not to answer"))]

sns.heatmap(all_answered_norm.pivot_table(
    index="Age", columns=["Gender"],
    values="Salary Delta", aggfunc=np.mean))

<Axes: xlabel='Gender', ylabel='Age'>


sns.heatmap(salaries_df.pivot_table(
    index="Age", columns=["Education"],
    values="Salary Delta", aggfunc=np.mean))

<Axes: xlabel='Education', ylabel='Age'>


salaries_df[(salaries_df["Age"] == "under 18") & (salaries_df["Education"] == "PhD")]


sns.heatmap(all_answered_norm.pivot_table(
    index="Education", columns=["Gender"],
    values="Salary Delta", aggfunc=np.mean))

<Axes: xlabel='Gender', ylabel='Education'>

Education	College degree	High School	Master's degree	PhD	Professional degree (MD, JD, etc.)	Some college	All
Gender
Man	0.095	0.007	0.047	0.008	0.007	0.020	0.184
Non-binary	0.014	0.001	0.007	0.001	0.001	0.003	0.026
Other or prefer not to answer	0.005	0.000	0.003	0.000	0.000	0.001	0.010
Prefer not to answer	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Woman	0.378	0.010	0.266	0.038	0.041	0.047	0.780
All	0.491	0.018	0.323	0.048	0.049	0.071	1.000

	title	avg_salary	location	formatted_experience_level
0	lead solar maintenance roofer, residential ser...	67600.000	San Diego, CA	Mid-Senior level
1	mental health professional	67410.500	Spokane, WA	Entry level
2	mental health professional	67410.500	Spokane, WA	Entry level
3	clinic managers - physical therapist	104260.650	Puyallup, WA	Mid-Senior level
4	territory associate	47500.000	San Gabriel, CA	Entry level
...	...	...	...	...
3112	equipment maintenance technician- automotive p...	52000.000	Dayton, OH	Mid-Senior level
3113	senior analyst	80000.000	New York, NY	Mid-Senior level
3114	design intern	44688.800	Miramar, FL	Mid-Senior level
3115	sr. bilingual copywriter (canadian french)	98550.000	Torrance, CA	Mid-Senior level
3116	mri manager	122500.000	New York, NY	Mid-Senior level

	title	location	formatted_experience_level
0	research and instruction librarian	Boston, MA	Associate
1	marketing specialist	Chattanooga, TN	Entry level
2	program manager	Milwaukee, WI	Associate
3	accounting manager	Greenville, SC	Associate
4	scholarly publishing librarian	Hanover, NH	Entry level
...	...	...	...
22442	devops engineer	Jersey CIty, NJ	Entry level
22443	tutor	bristol, RI	Entry level
22444	it manager	Cincinnati, OH	Entry level
22445	senior engineer i	Ithaca, NY	Associate
22446	group copy supervisor	New York City, NY	Mid-Senior level

	0
0	84594.480
1	81903.760
2	94990.000
3	87707.500
4	78020.000
...	...
22442	113870.000
22443	78020.000
22444	93755.000
22445	124620.000
22446	89856.000

	Job Title	Location	Mapped Experience (In Field)	Age	Gender	Education	Predicted Salary	Annual Salary	Salary Delta
5470	program manager	Austin, TX	Associate	35-44	Woman	PhD	94990.000	95000.000	10.000
16450	talent acquisition advisor	Seattle, WA	Director	45-54	Man	College degree	93788.000	93800.000	12.000
19559	program management department manager	Novi, MI	Director	35-44	Woman	College degree	99837.650	99850.000	12.350
4841	communications specialist	Washington, DC	Associate	35-44	Woman	Master's degree	93016.750	93000.000	16.750
16342	office manager	Greensboro, NC	Executive	55-64	Woman	Some college	67982.000	68000.000	18.000
...	...	...	...	...	...	...	...	...	...
7659	senior policy advisor	DC, DC	Director	35-44	Woman	PhD	84760.000	1334782.000	1250022.000
5666	principal software engineer	Boston, MA	Associate	25-34	Man	Master's degree	149470.000	1650000.000	1500530.000
4792	attending physician (general internal medicine)	New Haven, CT	Associate	25-34	Woman	Professional degree (MD, JD, etc.)	110419.800	1900000.000	1789580.200
1712	owner and ceo	New York , NY	Director	55-64	Woman	Master's degree	90144.000	3000000.000	2909856.000
21275	inside sales manager	Pinebrook, NJ	Director	55-64	Woman	Master's degree	77476.100	5000044.000	4922567.900

	job_id	company_id	title	description	max_salary	med_salary	min_salary	pay_period	formatted_work_type	location	applies	original_listed_time	remote_allowed	views	job_posting_url	application_url	application_type	expiry	closed_time	formatted_experience_level	skills_desc	listed_time	posting_domain	sponsored	work_type	currency	compensation_type
0	85008768	NaN	Licensed Insurance Agent	While many industries were hurt by the last fe...	52000.0	NaN	45760.0	YEARLY	Full-time	Chico, CA	NaN	1.690000e+12	NaN	5.0	https://www.linkedin.com/jobs/view/85008768/?t...	NaN	ComplexOnsiteApply	1.710000e+12	NaN	NaN	NaN	1.690000e+12	NaN	1	FULL_TIME	USD	BASE_SALARY
1	133114754	77766802.0	Sales Manager	Are you a dynamic and creative marketing profe...	NaN	NaN	NaN	NaN	Full-time	Santa Clarita, CA	NaN	1.690000e+12	NaN	NaN	https://www.linkedin.com/jobs/view/133114754/?...	NaN	ComplexOnsiteApply	1.700000e+12	NaN	NaN	NaN	1.690000e+12	NaN	0	FULL_TIME	NaN	NaN
2	133196985	1089558.0	Model Risk Auditor	Join Us as a Model Risk Auditor – Showcase You...	NaN	NaN	NaN	NaN	Contract	New York, NY	1.0	1.690000e+12	NaN	17.0	https://www.linkedin.com/jobs/view/133196985/?...	NaN	ComplexOnsiteApply	1.700000e+12	NaN	NaN	NaN	1.690000e+12	NaN	0	CONTRACT	NaN	NaN
3	381055942	96654609.0	Business Manager	Business ManagerFirst Baptist Church ForneyFor...	NaN	NaN	NaN	NaN	Full-time	Forney, TX	NaN	1.690000e+12	NaN	NaN	https://www.linkedin.com/jobs/view/381055942/?...	NaN	ComplexOnsiteApply	1.700000e+12	NaN	NaN	NaN	1.690000e+12	NaN	0	FULL_TIME	NaN	NaN
4	529257371	1244539.0	NY Studio Assistant	YOU COULD BE ONE OF THE MAGIC MAKERS\nKen Fulk...	NaN	NaN	NaN	NaN	Full-time	New York, NY	NaN	1.690000e+12	NaN	2.0	https://www.linkedin.com/jobs/view/529257371/?...	NaN	ComplexOnsiteApply	1.710000e+12	NaN	NaN	NaN	1.690000e+12	NaN	1	FULL_TIME	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
15881	3701373516	74718032.0	Sanitation Technician	Location:\n\nWest Columbia, SC, US, 29172\n\n2...	NaN	NaN	NaN	NaN	Part-time	West Columbia, SC	NaN	1.690000e+12	NaN	1.0	https://www.linkedin.com/jobs/view/3701373516/...	https://aspirebakeriescareers.com/job/West-Col...	OffsiteApply	1.700000e+12	NaN	Entry level	NaN	1.690000e+12	aspirebakeriescareers.com	0	PART_TIME	NaN	NaN
15882	3701373522	38897.0	Unit Secretary	Job Title: Unit Secretary\nDepartment: Nursing...	NaN	NaN	NaN	NaN	Full-time	Teaneck, NJ	2.0	1.690000e+12	NaN	7.0	https://www.linkedin.com/jobs/view/3701373522/...	https://recruiting.ultipro.com/HOL1005HNMC/Job...	OffsiteApply	1.700000e+12	NaN	Entry level	NaN	1.690000e+12	recruiting.ultipro.com	0	FULL_TIME	NaN	NaN
15883	3701373523	38897.0	Radiology Aide, Perdiem	Job Title: Radiology Aide, Perdiem\nDepartment...	NaN	NaN	NaN	NaN	Part-time	Teaneck, NJ	NaN	1.690000e+12	NaN	3.0	https://www.linkedin.com/jobs/view/3701373523/...	https://recruiting.ultipro.com/HOL1005HNMC/Job...	OffsiteApply	1.700000e+12	NaN	Entry level	NaN	1.690000e+12	recruiting.ultipro.com	0	PART_TIME	NaN	NaN
15884	3701373524	2623.0	MRI Manager	Grade 105\nJob Type: Officer of Administration...	135000.0	NaN	110000.0	YEARLY	Full-time	New York, NY	NaN	1.690000e+12	NaN	10.0	https://www.linkedin.com/jobs/view/3701373524/...	https://opportunities.columbia.edu/jobs/mri-ma...	OffsiteApply	1.700000e+12	NaN	Mid-Senior level	NaN	1.690000e+12	opportunities.columbia.edu	0	FULL_TIME	USD	BASE_SALARY
15885	3701373527	84659.0	Area Director of Business Development	Nexion Health Management affiliates operate 56...	NaN	NaN	NaN	NaN	Full-time	Vicksburg, MS	2.0	1.690000e+12	NaN	31.0	https://www.linkedin.com/jobs/view/3701373527/...	NaN	ComplexOnsiteApply	1.700000e+12	NaN	NaN	NaN	1.690000e+12	NaN	0	FULL_TIME	NaN	NaN

	Timestamp	How old are you?	What industry do you work in?	Job title	If your job title needs additional context, please clarify here:	What is your annual salary? (You'll indicate the currency in a later question. If you are part-time or hourly, please enter an annualized equivalent -- what you would earn if you worked the job 40 hours a week, 52 weeks a year.)	How much additional monetary compensation do you get, if any (for example, bonuses or overtime in an average year)? Please only include monetary compensation here, not the value of benefits.	Please indicate the currency	If "Other," please indicate the currency here:	If your income needs additional context, please provide it here:	What country do you work in?	If you're in the U.S., what state do you work in?	What city do you work in?	How many years of professional work experience do you have overall?	How many years of professional work experience do you have in your field?	What is your highest level of education completed?	What is your gender?	What is your race? (Choose all that apply.)
0	4/27/2021 11:02:10	25-34	Education (Higher Education)	Research and Instruction Librarian	NaN	55,000	0.0	USD	NaN	NaN	United States	Massachusetts	Boston	5-7 years	5-7 years	Master's degree	Woman	White
1	4/27/2021 11:02:22	25-34	Computing or Tech	Change & Internal Communications Manager	NaN	54,600	4000.0	GBP	NaN	NaN	United Kingdom	NaN	Cambridge	8 - 10 years	5-7 years	College degree	Non-binary	White
2	4/27/2021 11:02:38	25-34	Accounting, Banking & Finance	Marketing Specialist	NaN	34,000	NaN	USD	NaN	NaN	US	Tennessee	Chattanooga	2 - 4 years	2 - 4 years	College degree	Woman	White
3	4/27/2021 11:02:41	25-34	Nonprofits	Program Manager	NaN	62,000	3000.0	USD	NaN	NaN	USA	Wisconsin	Milwaukee	8 - 10 years	5-7 years	College degree	Woman	White
4	4/27/2021 11:02:42	25-34	Accounting, Banking & Finance	Accounting Manager	NaN	60,000	7000.0	USD	NaN	NaN	US	South Carolina	Greenville	8 - 10 years	5-7 years	College degree	Woman	White
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
27935	2/10/2023 16:15:00	25-34	Oil & Gas	Finance process Expert	NaN	25000	0.0	USD	NaN	NaN	Colombia	NaN	Bogota	8 - 10 years	8 - 10 years	Master's degree	Man	Hispanic, Latino, or Spanish origin
27936	2/16/2023 19:14:37	25-34	Computing or Tech	It manager	NaN	55000	NaN	USD	NaN	NaN	United States	Ohio	Cincinnati	11 - 20 years	2 - 4 years	Some college	Man	White
27937	2/16/2023 20:11:48	25-34	Engineering or Manufacturing	Senior Engineer I	Automotive mechanical engineering	87000	7000.0	USD	NaN	NaN	United States	New York	Ithaca	5-7 years	5-7 years	College degree	Man	White
27938	2/17/2023 17:58:48	25-34	Computing or Tech	software engineer	NaN	64000	0.0	USD	NaN	NaN	denmark	NaN	århus	2 - 4 years	2 - 4 years	College degree	Man	White
27939	2/21/2023 13:54:14	35-44	Marketing, Advertising & PR	Group Copy Supervisor	NaN	150000	NaN	USD	NaN	NaN	US	New York	New York City	11 - 20 years	8 - 10 years	Master's degree	Woman	White

Education	College degree	High School	Master's degree	PhD	Professional degree (MD, JD, etc.)	Some college	All
Gender
Man	2136	155	1064	189	155	448	4147
Non-binary	306	12	169	16	14	72	589
Other or prefer not to answer	114	7	62	10	9	17	219
Prefer not to answer	0	0	0	1	0	0	1
Woman	8514	222	5993	867	929	1071	17596
All	11070	396	7288	1083	1107	1608	22552

The Relationship Between Demographic Information and the Difference Between Expected and Actual Salary¶