Suhani Bhatia
← Work

Independent analysis · 2026

From a public table to a usable file

Independent technical work · Python · SQL · reproducible extract

Python and SQL are listed as beginner skills on my résumé. This page treats them at that scale. The task is modest: take published NFHS-5 Phase-II totals, check them, compute a phone gap, and write a clean extract. The scripts in the artefacts folder do that.

Raw input: nfhs_phase2.csv. Cleaning: types, range checks (percentages between 0 and 100), a uniqueness check on state codes. SQL: order states by bank - mobile. Python: the same join in pandas, a CSV out. Chart: the ranked map on the mobile-geography page. Implication: a digital welfare channel should be costed against the phone number the system can actually reach.

clean.pyquery.sqlInput CSV