Digital Development Convergence in the United States: A Spatial and Temporal Analysis from 2013 to 2023

This paper won the Economic Writing Competition, 2025 fall (Top 1 paper).

Access the R shiny web app through: https://harrylu0708.shinyapps.io/digital-convergence/.

This study investigates the convergence of digital infrastructure across U.S. counties and states from 2013 to 2023, with a focus on whether regional disparities in digital access have narrowed over time. Using microdata from the Integrated Public Use Microdata Series (IPUMS USA), we construct a household-level digital access index based on device and internet connectivity variables. We apply both standard σ- and β-convergence models as well as a spatial lag β-convergence model to evaluate two key hypotheses: (1) initially disadvantaged regions have experienced faster digital growth, and (2) geographically proximate regions converge together through spatial spillovers. Our results provide strong evidence of both σ- and β-convergence at the county and state levels, indicating a general reduction in the digital divide. However, we find no robust support for spatial spillovers—regions do not appear to converge faster as a result of their neighbors' growth. This suggests that while local catch-up dynamics are at play, geographic diffusion of digital development may be limited by policy fragmentation, administrative boundaries, or infrastructure barriers.

Reproducibility Instructions

Prerequisites

• R Version: R 4.0 or higher is recommended
• RStudio: Optional but recommended for easier navigation

Step 1: Install Required R Packages

Install all required R packages by running the following in your R console:

# Core data manipulation and visualization packages
install.packages(c("dplyr", "tidyr", "readr", "ggplot2", "ggrepel"))

# IPUMS data access package
install.packages("ipumsr", type = "source")

# Spatial analysis packages
install.packages(c("spdep", "sf", "spatialreg"))

# Convergence analysis package
install.packages("ConvergenceClubs")

# Shiny app packages (for interactive visualization)
install.packages(c("shiny", "shinydashboard", "DT", "plotly"))

Alternatively, you can install all packages at once:

required_packages <- c("dplyr", "tidyr", "readr", "ggplot2", "ggrepel", 
                      "ipumsr", "spdep", "sf", "spatialreg", "ConvergenceClubs",
                      "shiny", "shinydashboard", "DT", "plotly")

install.packages(required_packages)

Step 2: Data Setup

1. IPUMS USA Data:

   • The project uses IPUMS USA microdata files (usa_00005.xml and usa_00006.dat)
   • These files should be placed in the data/raw_data/ directory
   • If you need to download the data, visit IPUMS USA and extract the DDI file and microdata file

2. Geographic Data:

   • County shapefiles are already included in data/geo_data/
   • Files: cb_2013_us_county_500k.shp and associated files

3. Directory Structure: Ensure your project directory structure matches:

   Digital-Drifting/
   ├── data/
   │   ├── raw_data/          # IPUMS data files
   │   ├── geo_data/          # Shapefiles
   │   ├── cleaned_data/      # Processed data (generated)
   │   └── results/           # Analysis results (generated)
   ├── data_processing/
   ├── convergence/
   ├── robustness_analysis/
   └── images/                # Generated plots
   

Step 3: Set Working Directory

Before running any scripts, set your working directory to the project root:

setwd("/path/to/Digital-Drifting")

Or open the writing_project.Rproj file in RStudio, which will automatically set the working directory.

Step 4: Execute Scripts in Order

1. Data Processing (data_processing/processing.R):

   • Processes raw IPUMS data
   • Creates digital access index
   • Generates cleaned datasets in data/cleaned_data/

2. Convergence Analysis:

   • convergence/sigma_convergence.R - σ-convergence analysis
   • convergence/beta_convergence.R - County-level β-convergence
   • convergence/beta_convergence_state_level.R - State-level β-convergence
   • convergence/spatial_convergence.R - Spatial lag β-convergence
   • convergence/club_convergence.R - Club convergence analysis

3. Robustness Checks (robustness_analysis/):

   • robust_analysis_beta.R - Robustness checks for β-convergence
   • robust_analysis_sigma.R - Robustness checks for σ-convergence
   • min_index.R - Analysis with minimum index
   • pca_index.R - Analysis with PCA index

Step 5: Expected Outputs

After running the scripts, you should see:

• Cleaned data files in data/cleaned_data/
• Results files in data/results/
• Visualization plots in images/

Step 6: Interactive Visualization with Shiny App

An interactive R Shiny web application is available to visualize the convergence analysis results.

To run the Shiny app:

1. Ensure you have completed the data processing steps (Steps 1-4) and have the cleaned data files available.

2. Launch the app using one of the following methods:

   Method 1: From R Console

   library(shiny)
   runApp("app.R")

   Method 2: From RStudio

   • Open app.R in RStudio
   • Click the "Run App" button in the top-right corner of the editor

   Method 3: From Command Line

   Rscript -e "shiny::runApp('app.R')"

3. The app will open in your default web browser. The dashboard includes:

   • Overview: Summary statistics and project information
   • Sigma Convergence: Interactive time series visualization of σ-convergence
   • Beta Convergence - County: County-level β-convergence scatter plots with region type filters
   • Beta Convergence - State: State-level β-convergence analysis
   • Data Explorer: Interactive data table with filtering capabilities
   • Summary Statistics: Descriptive statistics and distribution plots

Features:

• Interactive plots using Plotly (zoom, pan, hover for details)
• Filterable data tables
• Real-time model summaries
• Region type filtering for county-level analysis

Notes

• Some scripts may require manual adjustment of file paths depending on your system
• The IPUMS data files are large and may take time to process
• Spatial analysis scripts require the geographic data files to be properly loaded
• If you encounter package installation issues, ensure you have the latest version of R and try installing packages individually
• The Shiny app requires the cleaned data files (data/cleaned_data/data_with_county.csv) to be present. If these files don't exist, run the data processing scripts first.