Date of Award

2026

Document Type

Thesis

Degree Name

Master of Science in Artificial Intelligence

Department

Digital Engineering

Committee Chair and Members

Naoual Amrouche

Keywords

Arts and culture, Data envelopment analysis, Efficiency measuement, GDP, Machine learning, US state-level analysis

Abstract

Arts and cultural production contributed $1.17 trillion to United States gross domestic product in 2023, 4.2% of the national total, but that contribution is spread very unevenly across states, and the official statistics describe how large the sector is rather than how efficiently it operates (Bureau of Economic Analysis, 2024). This study asks how efficiently each state convert growth in arts and culture employment into growth in value added, how many inputs the efficiency model can carry before it stops distinguishing 51 observations, and whether observable state characteristics explain the differences found. The analysis uses 2023 state-level data for all 50 states and the District of Columbia from the Bureau of Economic Analysis Arts and Cultural Production Satellite Account, with demographic, income, tourism, and arts funding data from the Census Bureau, the International Trade Administration, and the National Assembly of State Arts Agencies. The first stage estimates output-oriented Data Envelopment Analysis under variable returns to scale. The second stage treats the resulting efficiency scores as a dependent variable and models them on 13 and 12 state-level predictors using Ordinary Least Squares, Random Forest, Extreme Gradient Boosting, a Generalized Additive Model, Multivariate Adaptive Regression Splines, and a Gamma Generalized Linear Model. The study demonstrates that arts and culture efficiency can be meaningfully measured using DEA approach, although results depend on the specification used. The second-stage analysis serves primarily as an explanatory framework, with its strength lying in identifying potential linear and non-linear relationships rather than making predictions. Across the models, however, no single state-level characteristic consistently explains differences in efficiency. The findings highlight both the usefulness of efficiency measurement and the importance of specification choices when interpreting the factors associated with observed differences. Keywords: Efficiency Measurement, Data Envelopment Analysis, Arts and Culture, ACPSA Dataset, US State-level Analysis, Explanatory Modeling, Machine Learning, GDP

Download

Share

COinS