Background

The TIC Domicílios survey is conducted annually by CETIC.br, the Regional Center for Studies on the Development of the Information Society at NIC.br, to measure ICT access and use across Brazilian households. The survey uses a complex multi-stage probability design with bootstrap replicate weights published by the producer.

What the analysis covers

The comparison reproduces a representative set of TIC Domicílios indicators in both R and Python, focusing on:

• Bootstrap variance estimation, including the distinction between the Canty-Davison scale convention used by R’s survey (extracted from repdesign$scale after replicate construction) and the Rao-Wu scale convention used internally by svy.
• Proportion estimation, comparing the Taylor-linearized indicator approach (svymean) with the logit-scale approach (svyciprop) recommended for proportions near 0 or 1.
• Domain estimation by geographic area.

After aligning the bootstrap scale convention, the two implementations agree to machine precision.

Why this matters

TIC Domicílios is a production national survey at a producer-set bootstrap scale, not a textbook example. Confirming that svy reproduces R’s results on real producer microdata — once the variance conventions are made explicit — is the right kind of stress test for an analytical package. The exercise also surfaces a practical point for users: producer-published replicate weights carry scale information that downstream tools must respect.

Read the full post

→ Comparação entre survey (R) e svy (Python) — Thiago Meireles

Background

The Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH) is Mexico’s national income and expenditure survey, conducted by INEGI. It uses a complex stratified multi-stage probability design and is the primary microdata source for measuring poverty, inequality, and household consumption in Mexico.

What the analysis covers

The replication walks through a full ENIGH analytical workflow in Python and contrasts it with the equivalent srvyr code in R. It covers:

• Survey design declaration with ENIGH’s stratification, primary sampling units, and final weights.
• Weighted point estimation and standard errors for indicators including the housing-deprivation indicator (rezago habitacional).
• Disaggregation by federal entity, demonstrating domain estimation across Mexico’s 32 states.

The Python and R implementations produce numerically equivalent results across the indicators tested.

Why this matters

ENIGH is widely used by Mexican researchers, INEGI, CONEVAL, and consultancies for poverty and inequality analysis. A reproducible Python pathway that matches the established R workflow lowers the barrier for analysts who work primarily in Python to engage with ENIGH microdata without changing their existing tooling.

Read the full post

→ De R a Python: trabajar con microdatos de encuestas complejas — Claudio Pacheco