3. Data
Purpose
Describe the data sources and datasets used in this research.
Details
Dataset Overview: The primary dataset is a single flat CSV file containing N=10,000 synthetically generated patient outpatient records.
Source: The data was simulated using a custom Python script for the express purpose of demonstrating this research workflow. It does not represent any specific institution, clinic, or geography.
Pre-processing: - lead_time_days was bounded between 1 and 365. - deprivation_quintile was uniformly distributed. - The outcome variable missed_appointment was synthetically weighted so that higher previous_missed and higher lead_time_days slightly increase the probability of a missed visit. Overall class balance is approximately 80% attended, 20% missed.
Applied Practice & Innovation: An AI coding assistant was used to rapidly generate the Python script responsible for synthesizing this dataset. The prompt specified the distribution constraints and correlation structures, which were then reviewed and tested by human researchers before final synthesis.