# Study sessions reproducibility package

## What this reproduces

The project calculates mean study-session duration for two sound settings. The four records are synthetic and illustrate the workflow.

## Files

| Path | Purpose |
| --- | --- |
| `data/sessions.csv` | Input data. |
| `data/data-dictionary.md` | Field definitions, units and missing-value rules. |
| `analysis/analyze.py` | Groups sessions by sound setting and calculates the mean duration. |
| `expected/means.csv` | Reference table for checking the run. |
| `results/means.csv` | Generated output after the script runs. |

## Run and verify

From the extracted project folder, run:

```sh
python --version
python analysis/analyze.py
```

If your system uses `python3` or `py -3` for Python 3, use that command instead of `python`. The example was tested on Linux x86_64 with Python 3.12.14 and uses only the Python standard library.

Open `results/means.csv` and compare it with `expected/means.csv`. Background noise should have a mean of 20.0 minutes and quiet should have a mean of 25.0 minutes. Running the script again replaces the generated table.

## Adapt this README to your study

Replace the explanation and file list with your own. Include one starting command, the exact interpreter and dependency versions you tested, an input dataset version or access instructions, the code revision, expected outputs, and the applicable data and code licenses. If readers can only run a demonstration with synthetic data, explain which inputs and results differ from the study.
