The Global Data Jobs Analysis project is an end-to-end exploratory data analysis (EDA) and visualization study of the global data job market. The objective of this project is to extract meaningful insights from real-world job posting data and understand how factors such as role, salary, location, skills, and work type influence compensation and hiring trends.
The project combines Python-based data analysis with Power BI dashboards to deliver both statistical rigor and business-oriented insights. Using Python, the dataset was cleaned, transformed, and analyzed to uncover patterns in salary distribution, missing data behavior, and variability across roles. Key statistical techniques such as skewness analysis, standard deviation, coefficient of variation, and quantile-based segmentation were applied to better understand the structure of the data.
On the visualization side, Power BI was used to build interactive dashboards that highlight:
- Global demand for different data roles
- Salary comparisons across job titles and countries
- Hiring trends over time
- Skill and education requirements
- Remote work and employment patterns
GLOBAL DATA JOBS ANALYSIS/
│
├── data/
│ └── global_jobs_data.xlsx # Dataset
│
|── power_bi/ # Data Visualization
| └── datajobs_visualization.pbit
| └── visualization.pdf
│ └── job_analysis_images/
| └── data_tables.jpg
| └── jobs_density.jpg
| └── data_jobs_percent_analysis.jpg
| └── jobs_median_salary.jpg
| └── key_performance_indicator.jpg
| └── quarterly_matrix.jpg
| └── trend_analysis.jpg
|
├── datajobs_venv/ # Virtual environment (not tracked)
├── datajobs_analysis.ipynb # Main analysis notebook
├── .gitignore
├── requirements.txt
├── README.md
├── LICENSE
Follow these steps to run the project locally:
git clone https://github.com/WizardNox/Global-Data-Jobs-Analysis-and-Visualization.git
cd data-jobs-edapython -m venv datajobs_venv- Windows:
datajobs_venv\Scripts\activate- Mac/Linux:
source datajobs_venv/bin/activatepip install -r requirements.txtOpen the Jupyter Notebook:
jupyter notebookThen navigate to the main notebook file and run all cells.
- Ensure the dataset is located inside the
data/folder - File paths in the notebook are configured using relative paths for portability
- All required libraries are listed in
requirements.txt
Power BI was used as the primary visualization and business intelligence tool in this project. While Python handled statistical validation and data preparation, Power BI enabled interactive exploration, trend analysis, and decision-focused insights.
Below are the key dashboards and insights derived:
- Data Engineer roles dominate hiring volume, followed by Data Analyst and Data Scientist roles
- Despite lower volume, Cloud Engineers and Machine Learning Engineers command higher salaries
- Senior roles show higher compensation but lower job counts, indicating specialization
- The market favors data infrastructure roles (Data Engineering)
- High-paying roles are skill-intensive and less saturated
- Senior Data Scientist ($156K) and Machine Learning Engineer ($155K) are the top-paying roles
- Followed by Senior Data Engineer and Software Engineer roles
- Entry-level roles like Data Analyst (~$90K) are significantly lower
- Countries like the USA, UK, and Germany offer higher salaries
- India shows lower salary levels, but similar role hierarchy
- Senior Data Engineer and Data Scientist roles dominate globally
- Compensation strongly correlates with Experience(Senior roles),Technical depth(Machine Learning,Engineering)
- Clear geographic salary disparity
- Supports global hiring cost strategies
- Hiring peaks during July–September
- Hiring drops in November–December
- Trend consistent across roles
- Reflects seasonal hiring cycles
- End-of-year slowdown due to budget constraints
- ~33% of jobs do not require a degree
- Senior roles more likely to mention degrees
- ~87% jobs allow remote work, yet ~91% are full-time
- Shift toward skills-first hiring
- Remote work is normalized, but job structure remains traditional
High job density in:
- North America
- Europe
- India (emerging hub)
- Lower presence in developing regions
- Data jobs cluster in tech-driven economies
- Emerging markets show growth potential
- Data Engineering is the most in-demand role globally
- Senior roles command significantly higher salaries
- Strong geographic salary inequality exists
- Hiring follows seasonal patterns
- Industry is shifting toward skills over degrees
- Remote work is common, but full-time roles dominate
- Python
- Pandas – Data manipulation
- NumPy – Statistical operations
- Matplotlib / Seaborn – Visualization
- Power BI - Visualization
Key preprocessing steps performed:
- Converted
job_posted_dateto datetime format - Converted salary columns to numeric format
- Handled missing values in salary columns
- Removed rows with missing job titles
- Created clean subsets for statistical analysis
-
Identified overall salary distribution
-
Observed:
- Mean salary ≈ $120K
- Median salary ≈ $113K
- Maximum salary ≈ $920K
Insight:
- The dataset shows a right-skewed distribution
- A few extremely high salaries pull the average upward
Used Interquartile Range (IQR) to detect outliers:
- Q1 (25th percentile)
- Q3 (75th percentile)
- IQR = Q3 - Q1
Outliers defined as:
Lower Bound = Q1 - 1.5 * IQR
Upper Bound = Q3 + 1.5 * IQR
Insight:
- Significant high-end salary outliers exist
- Median is a more reliable measure than mean in skewed distributions
Used pd.qcut() to divide salaries into 4 equal groups:
- Low
- Mid
- High
- Very High
Insight:
- Each band contains ~25% of data
- Demonstrates balanced segmentation using quantiles
Insight:
- Observed salary differences are statistically significant
- Not due to random chance
Analyzed:
- Standard deviation
- Skewness
- Spread of salaries
Insight:
- High standard deviation indicates wide salary variation
- Outliers significantly impact distribution
- Salary distributions are right-skewed with high-end outliers
- Entry-level roles tend to have compressed salary bands
- Senior roles show higher variability in compensation
- Geographic location significantly impacts compensation
- Strong data cleaning and wrangling skills
- Ability to apply statistical reasoning in real datasets
- Understanding of data distributions and outliers
- Capability to extract business insights from raw data
- Experience with real-world messy datasets
This project is part of my journey to becoming a highly skilled data analyst. It focuses not just on coding, but on thinking like an analyst — questioning data, validating assumptions, and extracting real-world insights.
This project is licensed under the terms of the LICENSE file included in this repository.