최신 데이터사이언스 연봉 데이터

2020년~2023년데이터 사이언스 분야의 보상 추세 및 변동에 대한 데이터

  • 직업 이름(문자열)
  • 고용 유형(문자열) - 풀타임,프리랜서,파트타임
  • 경험 수준(문자열) - Entry,Mid,Senoir,Executive
  • 전문 지식수준(문자열) - Junior, Intermediate, Dicrector,
  • 급여(정수)
  • 급여 통화(문자열)- United States, Dollar,Euro 등
  • 회사위치(문자열) - United States,United Kingdom 등
  • USD급여(정수)
  • 직원 거주지(문자열) - United States, Germany 등
  • 회사의 규모(문자열) - Small, Medium, Large
  • 연도(정수) - 2020, 2021, 2022, 2023
In [18]:
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df = pd.read_csv("C:/Users/lucy8/Downloads/Latest_Data_Science_Salaries.csv")
df_23 = df[df['Year'] == 2023]
In [19]:
print(df_23.shape)
print(df_23.info())
print(df_23.isnull().sum())

(1996, 11)
<class 'pandas.core.frame.DataFrame'>
Int64Index: 1996 entries, 0 to 2036
Data columns (total 11 columns):
 #   Column              Non-Null Count  Dtype 
---  ------              --------------  ----- 
 0   Job Title           1996 non-null   object
 1   Employment Type     1996 non-null   object
 2   Experience Level    1996 non-null   object
 3   Expertise Level     1996 non-null   object
 4   Salary              1996 non-null   int64 
 5   Salary Currency     1996 non-null   object
 6   Company Location    1996 non-null   object
 7   Salary in USD       1996 non-null   int64 
 8   Employee Residence  1996 non-null   object
 9   Company Size        1996 non-null   object
 10  Year                1996 non-null   int64 
dtypes: int64(3), object(8)
memory usage: 187.1+ KB
None
Job Title             0
Employment Type       0
Experience Level      0
Expertise Level       0
Salary                0
Salary Currency       0
Company Location      0
Salary in USD         0
Employee Residence    0
Company Size          0
Year                  0
dtype: int64
In [20]:
df_23
Out[20]:
Job Title Employment Type Experience Level Expertise Level Salary Salary Currency Company Location Salary in USD Employee Residence Company Size Year
0 Data Engineer Full-Time Senior Expert 210000 United States Dollar United States 210000 United States Medium 2023
1 Data Engineer Full-Time Senior Expert 165000 United States Dollar United States 165000 United States Medium 2023
2 Data Engineer Full-Time Senior Expert 185900 United States Dollar United States 185900 United States Medium 2023
3 Data Engineer Full-Time Senior Expert 129300 United States Dollar United States 129300 United States Medium 2023
4 Data Scientist Full-Time Senior Expert 140000 United States Dollar United States 140000 United States Medium 2023
... ... ... ... ... ... ... ... ... ... ... ...
2032 Machine Learning Engineer Full-Time Mid Intermediate 52000 British Pound Sterling United Kingdom 63980 United Kingdom Medium 2023
2033 Machine Learning Engineer Full-Time Mid Intermediate 48000 British Pound Sterling United Kingdom 59059 United Kingdom Medium 2023
2034 Machine Learning Engineer Full-Time Mid Intermediate 38000 British Pound Sterling United Kingdom 46755 United Kingdom Medium 2023
2035 Data Architect Full-Time Senior Expert 110000 United States Dollar United States 110000 United States Medium 2023
2036 Data Engineer Full-Time Senior Expert 236000 United States Dollar United States 236000 United States Medium 2023

1996 rows × 11 columns

In [4]:
plt.hist(df_23['Salary in USD'], bins=10, color='skyblue', edgecolor='black')
plt.title('Salary Distribution in USD')
plt.xlabel('USD')
plt.ylabel('Frequency')

plt.show()
In [5]:
unique_job_titles = df_23['Job Title'].unique()
num_unique_job_titles = len(unique_job_titles)

print(num_unique_job_titles)

plt.figure(figsize=(18, 6))
plt.bar(range(num_unique_job_titles), df_23['Job Title'].value_counts())
plt.xticks(range(num_unique_job_titles), unique_job_titles, rotation=90)
plt.xlabel('Job Title')
plt.ylabel('Count')
plt.title('Number of Unique Job Titles')
plt.tight_layout()
plt.show()

90
In [6]:
job_counts = df_23['Job Title'].value_counts()
print(job_counts)

Data Engineer                    413
Data Scientist                   353
Data Analyst                     282
Machine Learning Engineer        202
Analytics Engineer                85
                                ... 
Autonomous Vehicle Technician      1
Data Engineer 2                    1
Analytics Engineering Manager      1
AWS Data Architect                 1
Data Analytics Lead                1
Name: Job Title, Length: 90, dtype: int64
In [7]:
job_20 = job_counts.iloc[:19]

plt.figure(figsize=(10, 6))
sns.set_style('whitegrid')
sns.boxplot(x=df_23['Job Title'], y=df_23['Salary'], order=job_20.index)
plt.xlabel('Job Title')
plt.xticks(rotation=90)
plt.ylabel('Salary')
plt.title('Salary Distribution for Top 20 Job Titles in 2023')
plt.tight_layout()
plt.show()
In [8]:
job_15 = job_counts.iloc[:14]


plt.figure(figsize=(10, 6))
sns.set_style('whitegrid')

sns.boxplot(x=df_23['Job Title'], y=df_23['Salary'], order=job_15.index)
plt.xlabel('Job Title')
plt.xticks(rotation=90)
plt.ylabel('Salary')
plt.title('Salary Distribution for Top 15 Job Titles in 2023')
plt.tight_layout()
plt.show()
In [9]:
salary_stats_by_job = df.groupby('Job Title')['Salary in USD'].agg(['mean', 'median', 'max', 'min']).reset_index()
salary_stats_by_job
Out[9]:
Job Title mean median max min
0 AI Architect 237484.000000 209968.0 330000 200000
1 AI Developer 133177.777778 108000.0 275000 60000
2 AI Programmer 56021.750000 55000.0 74087 40000
3 AI Scientist 114117.333333 90919.0 417937 18053
4 AWS Data Architect 258000.000000 258000.0 258000 258000
... ... ... ... ... ...
106 Sales Data Analyst 60000.000000 60000.0 60000 60000
107 Software Data Engineer 111627.666667 74883.0 210000 50000
108 Staff Data Analyst 97499.000000 97499.0 179998 15000
109 Staff Data Scientist 134500.000000 134500.0 164000 105000
110 Staff Machine Learning Engineer 185000.000000 185000.0 185000 185000

111 rows × 5 columns

In [10]:
experience_salary_grouped = df_23.groupby('Experience Level')['Salary'].mean().reset_index()
expertise_salary_grouped = df_23.groupby('Expertise Level')['Salary'].mean().reset_index()

plt.figure(figsize=(12, 6))

plt.subplot(1, 2, 1)
sns.barplot(x='Experience Level', y='Salary', data=experience_salary_grouped)
plt.xlabel('Experience Level')
plt.ylabel('Salary (USD)')
plt.title('Average Salary by Experience Level')

plt.subplot(1, 2, 2)
sns.barplot(x='Expertise Level', y='Salary', data=expertise_salary_grouped)
plt.xlabel('Expertise Level')
plt.ylabel('Salary (USD)')
plt.title('Average Salary by Expertise Level')

plt.tight_layout()
plt.show()
In [11]:
import matplotlib.pyplot as plt
filtered_df = df_23.dropna(subset=['Employee Residence'])

residence_counts = filtered_df['Employee Residence'].value_counts()
total_unique_residences = len(residence_counts)

plt.figure(figsize=(12, 6))
residence_counts.plot(kind='bar')
plt.xlabel('Employee Residence')
plt.ylabel('Count')
plt.title(f'Number of Unique Employee Residences ({total_unique_residences} unique)')
plt.tight_layout()
plt.show()
In [12]:
filtered_df = df_23[df_23['Employee Residence'] != 'United States']

residence_counts = filtered_df['Employee Residence'].value_counts()
total_unique_residences = len(residence_counts)

plt.figure(figsize=(12, 6))
residence_counts.plot(kind='bar')
plt.xlabel('Employee Residence')
plt.ylabel('Count')
plt.title(f'Number of Unique Employee Residences excluding "United States" ({total_unique_residences} unique)')
plt.tight_layout()
plt.show()
In [13]:
filtered_df = df_23[(df_23['Employee Residence'] != 'United Kingdom') & (df_23['Employee Residence'] != 'United States')]

residence_counts = filtered_df['Employee Residence'].value_counts()
total_unique_residences = len(residence_counts)

plt.figure(figsize=(12, 6))
residence_counts.plot(kind='bar')
plt.xlabel('Employee Residence')
plt.ylabel('Count')
plt.title(f'Number of Unique Employee Residences excluding "United Kingdom" and "United States" ({total_unique_residences} unique)')
plt.tight_layout()
plt.show()
In [14]:
plt.figure(figsize=(10, 6))

sns.barplot(x='Company Size', y='Salary', data=df_23)
plt.xlabel('Company Size')
plt.ylabel('Salary (USD)')
plt.title('Average Salary by Company Size')
plt.tight_layout()
plt.show()
In [15]:
columns = ["Job Title", "Experience Level", "Expertise Level", "Salary in USD", "Company Size"]
df_23 = pd.DataFrame(df_23, columns=columns)

heatmap_data = df_23.pivot_table(index="Job Title", columns=["Experience Level", "Expertise Level"], values="Salary in USD")

plt.figure(figsize=(12, 20))
sns.heatmap(heatmap_data, annot=True, fmt=".0f", cmap="YlGnBu", linewidths=0.5)
plt.title("Salary in USD by Job Title, Experience Level, and Expertise Level")
plt.xlabel("Experience Level, Expertise Level")
plt.ylabel("Job Title")
plt.show()
In [16]:
from sklearn.preprocessing import LabelEncoder

label_encoder = LabelEncoder()
df_23['Job Title'] = label_encoder.fit_transform(df_23['Job Title'])
df_23['Experience Level'] = label_encoder.fit_transform(df_23['Experience Level'])
df_23['Expertise Level'] = label_encoder.fit_transform(df_23['Expertise Level'])
df_23['Company Size'] = label_encoder.fit_transform(df_23['Company Size'])

corr_matrix = df_23[['Job Title', 'Experience Level', 'Expertise Level', 'Salary in USD', 'Company Size']].corr()
plt.figure(figsize=(10, 8))
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', fmt='.2f', linewidths=0.5)
plt.title('Correlation Heatmap')
plt.show()
In [ ]: