#research-methods
2 articles
- Religiosity vs. GDP: A Regression That Passed Every Check — and a Causal Story It Can't Tell
A statistics assignment asked me to regress national GDP on a religiosity score. The model passed cleanly: a significant negative slope, ~38% of variance explained, both residual diagnostics clean. That cleanliness is exactly why it's dangerous — a passing model and a causal story are not the same thing.
- MGNREGA: The Regression That Held — and the State Where the Budget Line Goes to Zero
My graduate final project analyzed 740 Indian districts to test whether approved labor budget predicts rural women's paid workdays under MGNREGA. The correlation was strong (r ≈ .82–.86) and the diagnostics were honest. Then I looked at West Bengal, the one state whose funds were frozen — and had to decide exactly how much that chart was allowed to say.