Sports Motivation Analysis
An applied research analysis repository using simulated data to preserve privacy while keeping schema consistent for reproducible workflows.
Problem
Human-subject datasets require protection; sharing analysis code should not expose participant records.
Context and constraints
- Use simulated data with the same schema as the original study data
- Separate data generation from analysis
- Constraints: do not invent findings; keep placeholders if not documented
Approach
- Define analysis plan and schema
- Generate a simulated dataset preserving schema
- Run analysis and reporting in a reproducible structure
Architecture
flowchart LR
A[Study design] --> B[Define schema]
B --> C[Generate simulated dataset]
C --> D[Run analysis]
D --> E[Report + visuals]
E --> F[Privacy-safe sharing]
Implementation highlights
- Privacy-first data handling via simulated dataset with preserved schema
- Reproducible layout for notebooks/reports
- Clear boundaries for what is shared publicly
Results and impact
In terms of negatively impacting wellbeing, physical activity experiences should reduce the focus on outcome-oriented goals such as winning and losing, comparisons to peers, focusing on body image or appearance, and forced involvement in physical activity or sport.
Tech stack
Python, data analysis, reproducible reporting.
Links
- Repo: https://github.com/luyangsi/Sports_Motivitation_Analysis
- Projects index: /projects/
What I'd improve next
Add a documented analysis protocol and an automated notebook-to-HTML report pipeline in CI.