Simulation Code and Scenario Outputs for “Children’s Concert Programs as Cultural Investment: Skill Formation, Cultural Capital, and Design Benchmarks”
Description
This dataset contains the simulation code and generated outputs supporting the design-calibration exercise reported in the manuscript “Children’s Concert Programs as Cultural Investment: Skill Formation, Cultural Capital, and Design Benchmarks.” The materials are intended to support transparency, reproducibility, and future field-design work in cultural economics. The dataset does not contain observed child-level data, administrative attendance records, or empirical outcome data from the Enescu Festival. All individual-level files are simulated. The simulation implements a dual-output cultural-economics model in which children’s concert programs may affect two outcome families: a capability stock, including attention, executive function, and self-regulation, and a cultural-capital/future-demand stock, including repertoire familiarity, instrument recognition, venue comfort, reduced future participation costs, and willingness to attend future cultural events. The simulation is scenario-based rather than causal or empirical. It distinguishes four calibration regions: below-threshold/null, conservative short-program, central strong-implementation, and upper-bound sensitivity. These scenarios vary assumptions about exposure intensity, effective dose, pedagogical quality, family and school complements, age sensitivity, and crowd-out. The purpose is to identify the conditions under which positive effects would be plausible, the conditions under which null effects are expected, and the kinds of empirical field designs that would be required to test the framework. The repository includes the Python simulation script, simulated individual-level data for each scenario, scenario-summary outputs, group means, parameter-sensitivity results, cost-effectiveness grids, and generated figures. Cost-effectiveness values are illustrative and scenario-conditioned; they should not be interpreted as audited costs or as evidence that any specific program is cost-effective. These materials are best understood as supplementary resources for a conceptual and methodological paper. They provide a transparent computational implementation of the model and can be used by researchers, cultural institutions, or policy analysts to inspect assumptions, reproduce the scenario outputs, modify parameter values, or adapt the framework for future empirical evaluation of children’s cultural-participation programs.
Files
Steps to reproduce
1. Download and unzip the repository files into a local working directory. 2. Ensure that Python 3.9 or later is installed. The simulation uses standard scientific Python libraries: `numpy`, `pandas`, and `matplotlib`. 3. Install the required packages if they are not already available: `pip install numpy pandas matplotlib` 4. Open the main Python script: `jce_dual_output_simulation.py` 5. Check the user settings at the beginning of the script. By default, the script is set to `FAST_MODE=True`, which runs a lighter version of the simulation for quick reproduction. For more stable scenario medians, set: `FAST_MODE=False` 6. Run the script from the command line: `python jce_dual_output_simulation.py` 7. The script will generate an output folder containing the simulated individual-level datasets, scenario-summary file, group means, parameter-sensitivity results, cost-effectiveness grid, and figures. 8. Compare the newly generated files with the files provided in the repository. Minor numerical differences may occur if settings, random seed, software version, or `FAST_MODE` configuration are changed. 9. To modify the simulation, adjust the scenario parameters in the `make_scenarios()` function or the model parameters in the `ModelParams` class. The script can be adapted to test alternative assumptions about effective dose, pedagogical quality, family and school complements, crowd-out, age sensitivity, or cost-effectiveness. 10. The generated outputs should be interpreted as scenario-conditioned design benchmarks. They are not empirical estimates and do not represent observed Enescu Festival child-outcome data.