Data Analysis Workflow — Social Science Research

End-to-end data analysis: load, explore, analyze, and produce publication-ready output. Primary language is Python (transitioning from Stata); Stata workflows are also supported.

Input: Dataset path or description of analysis goal (e.g., "regress log wages on education with individual fixed effects using PSID panel 2000-2020").

Constraints

• Use relative paths everywhere — no hardcoded absolute paths
• Set seed once at top for all stochastic operations
• Save all scripts to scripts/ with descriptive names
• Save all outputs (figures, tables, processed data) to output/
• Follow coding-style.md for Python file structure

Workflow Phases