TeamOne Project

Overview

This project analyzes job creation and firm performance indicators using data from the ES-Indicators Database and New Comprehensive datasets.

Setup

Install dependencies:

pip install -r requirements.txt

Ensure the joined database exists:

python join_datasets.py

Files

join_datasets.py - Script to join the two Stata datasets and create DuckDB database
explore_database.py - Helper script to explore database structure and identify columns
dashboard.py - Static HTML dashboard generator
dashboard_streamlit.py - Interactive Streamlit dashboard
research_questions.md - Research questions and findings

Using the Dashboards

Option 1: Interactive Streamlit Dashboard (Recommended)

Run the interactive Streamlit dashboard:

streamlit run dashboard_streamlit.py

This will open a web browser with an interactive dashboard where you can: - Navigate between different research questions - Select columns dynamically - View visualizations - Explore the data with custom SQL queries

Option 2: Static HTML Dashboard

Generate static HTML visualizations:

python dashboard.py

This creates: - Individual chart HTML files in data_images/ - A combined dashboard.html file

Open dashboard.html in a web browser to view all visualizations.

Option 3: Explore Database Structure

To understand what columns are available in the database:

python explore_database.py

This will: - List all tables and columns - Search for columns matching research question keywords - Save a CSV file with all column names - Show sample data

Research Questions

See research_questions.md for detailed research questions and findings.

Key findings: 1. Company Size: Large companies create most jobs absolutely; medium companies create most relative to size 2. Firm Maturity: Negative correlation - newer firms create more jobs 3. Access to Credit: Significant indicator of job creation ability 4. Interest Rates: Unfavorable rates associated with employment decreases

Data

The joined database (data/joined_data.duckdb) contains: - joined_data - Main joined table - dataset1 - Original ES-Indicators dataset - dataset2 - Original New Comprehensive dataset

Datasets are linked via the idstd column.

Notes

Large data files (.dta files) are not committed to git (see .gitignore)
The DuckDB database is created locally and should be added to .gitignore if it’s large
All visualizations use Altair for interactive charts

# TeamOne Project ## Overview This project analyzes job creation and firm performance indicators using data from the ES-Indicators Database and New Comprehensive datasets. ## Setup 1. Install dependencies: ```bash pip install -r requirements.txt ``` 2. Ensure the joined database exists: ```bash python join_datasets.py ``` ## Files - `join_datasets.py` - Script to join the two Stata datasets and create DuckDB database - `explore_database.py` - Helper script to explore database structure and identify columns - `dashboard.py` - Static HTML dashboard generator - `dashboard_streamlit.py` - Interactive Streamlit dashboard - `research_questions.md` - Research questions and findings ## Using the Dashboards ### Option 1: Interactive Streamlit Dashboard (Recommended) Run the interactive Streamlit dashboard: ```bash streamlit run dashboard_streamlit.py ``` This will open a web browser with an interactive dashboard where you can: - Navigate between different research questions - Select columns dynamically - View visualizations - Explore the data with custom SQL queries ### Option 2: Static HTML Dashboard Generate static HTML visualizations: ```bash python dashboard.py ``` This creates: - Individual chart HTML files in `data_images/` - A combined `dashboard.html` file Open `dashboard.html` in a web browser to view all visualizations. ### Option 3: Explore Database Structure To understand what columns are available in the database: ```bash python explore_database.py ``` This will: - List all tables and columns - Search for columns matching research question keywords - Save a CSV file with all column names - Show sample data ## Research Questions See `research_questions.md` for detailed research questions and findings. Key findings: 1. **Company Size**: Large companies create most jobs absolutely; medium companies create most relative to size 2. **Firm Maturity**: Negative correlation - newer firms create more jobs 3. **Access to Credit**: Significant indicator of job creation ability 4. **Interest Rates**: Unfavorable rates associated with employment decreases ## Data The joined database (`data/joined_data.duckdb`) contains: - `joined_data` - Main joined table - `dataset1` - Original ES-Indicators dataset - `dataset2` - Original New Comprehensive dataset Datasets are linked via the `idstd` column. ## Notes - Large data files (`.dta` files) are not committed to git (see `.gitignore`) - The DuckDB database is created locally and should be added to `.gitignore` if it's large - All visualizations use Altair for interactive charts