Quality & Assurances · Inclusion Advocate · Non-Profit Aspirant
Surojit O
I publish test reports on myself — the work, and the honest self-audit beside it. Notes on what I'm learning, and on negotiating for clarity.
New Log
From Needs Assessment to Recommendation: The Belonging Fund ProjectMy team designed a fund to help students feel they belonged. Our biggest limitation was that we never spoke to a single student.
Past Logs
- 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.
- Precise About Forty, Silent About the Rest
A weekly discussion post on selection ethics and personnel law — not a graded deliverable, just the rawest writing in the whole archive — taught me that the law can be extraordinarily exact about one boundary while staying completely silent about another, and that noticing the gap is its own kind of quality control.
- 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.
- The QA Manager, Part IV: Rating the Raters — and Who Never Got a Vote
Closing out a four-part series on one role: designing the performance appraisal that would evaluate whoever gets hired into it taught me that a rating system is a value statement before it's a measurement tool — and mine confessed what I quietly reward, and whose voice I'd left out entirely.