"It's a Small World After All": A Critical Analysis of the 2024 Disney DAS Policy Shift.

Dataset and Code Repository for "It's a Small World After All": A Critical Analysis of the 2024 Disney DAS Policy Shift This repository contains survey data, Monte Carlo simulation code, visualizations, and analysis results supporting the manuscript submitted to Tourism Management. Contents Survey Data & Analysis DAS_DATA.xlsx - Anonymized survey responses (n=269) from prior DAS users post-July 2024 policy change DAS_Analysis_Results.xlsx - Statistical analysis results including confidence intervals, effect sizes, and significance tests SURVEY.ipynb - Jupyter notebook containing complete survey analysis code (Python) Monte Carlo Simulation MONTECARLO.ipynb - Complete Monte Carlo simulation code (10,000 iterations, 6 scenarios, Python) mc_summary_table.csv - Main results: viability probabilities, mean/median net benefits, VaR(5%), costs mc_sensitivity_analysis.csv - Spearman rank correlations (ρ) between input parameters and net benefit outcomes Visualizations Survey Results (Figures 1-5): fig1_approval_and_impact.png - Approval rates and attendance impact breakdown fig2_anxiety_distribution.png - Anxiety level distribution (90.30% severe anxiety) fig3_key_metrics.png - Key metrics with 95% confidence intervals fig4_confidence_intervals.png - Statistical confidence intervals for primary outcomes fig5_approval_pathways.png - Application pathways and denial rates Monte Carlo Results: mc_net_benefit_distributions.png - Net benefit distributions across 6 scenarios (overlaid histograms) mc_cost_distributions.png - Cost structure comparison (box plots by scenario) mc_probability_positive.png - Viability probability (P(Net>0)) horizontal bar chart mc_sensitivity_tornado.png - Tornado diagram showing Spearman correlations (ρ) ranked by magnitude Key Findings Survey Evidence (n=269): 68.29% conditional denial rate among applicants 90.30% severe anxiety [95% CI: 86.16%-93.29%], NNH=1.24 85.13% attendance reduction [95% CI: 80.67%-89.22%] $2.13M estimated annual revenue loss per 500-family cohort Monte Carlo Results (10,000 iterations): Labor-intensive models fail: Baseline Professional 0.6% viable, -$4.38M mean loss Technology-mediated crisis support succeeds: 93.2% viable, +$1.48M mean benefit Cost differential: $6.46M → $646K (90% labor cost reduction) Sensitivity Analysis (Spearman Rank Correlations): Family spending: ρ=0.736 (dominates 54% of variance) Technology effectiveness: ρ=0.090 (explains <1% of variance) Fraud rate: ρ=-0.104 (economically irrelevant) Finding: Cost structure drives viability; operational parameters (fraud prevention, tech quality) barely matter Methodology Survey: Cross-sectional design, August-November 2024 Recruitment: IAAPA 2024 conference, disability advocacy networks 7-question instrument measuring denial rates, anxiety, behavioral intentions Statistical tests: Binomial tests, Cohen's h effect sizes, Number Needed to Harm Monte Carlo: 10,000 iterations with common random numbers (fair scenario comparison) 6 scenarios: Baseline Professional, Peer Volunteer, Technology-First, Premium Paid, Capacity-Limited, IBCCES-Verified Demand: Triangular distribution (66, 115, 165 families/day) Fraud rates: Beta distributions (standard 5%, IBCCES 2.5%) Tail risk: 15-21% annual probability (from 90% anxiety survey result) Convergence verified: 0.7% change n=7,500→10,000 Sensitivity Analysis: Spearman rank correlations between 9 input parameters and net benefit Robustness testing under worst-case scenarios (tech failure, high fraud, recession) Reproducibility All code is provided in Jupyter notebooks with: Complete parameter specifications Random seed (42) for reproducibility Inline documentation and comments Library versions specified Required Python libraries: NumPy 1.24+ SciPy 1.10+ Pandas 2.0+ Matplotlib 3.7+ Seaborn 0.12+ License Creative Commons Attribution 4.0 International (CC BY 4.0) You are free to: Share — copy and redistribute the material Adapt — remix, transform, and build upon the material Under the following terms: Attribution — You must give appropriate credit and indicate if changes were made Acknowledgments Survey respondents who shared experiences during a difficult policy transition. IAAPA 2024 conference organizers for facilitating data collection. Disability advocacy organizations for recruitment assistance. Data Availability Statement Survey data are anonymized to protect participant confidentiality. Raw individual responses are not included; only aggregate statistics and anonymized response patterns are provided. Monte Carlo simulation code is fully reproducible with provided random seed.