Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

fabrice-tiku-nyambod/data-decay-and-data-amnesia-what-a-survey-interruption-costs-global-health-surveillance-1

Domaine:

healthcare

Type de record:

paper
Créateur:
fab
Hôte:
Analysis pipeline for 'Data decay and data amnesia: what a survey interruption costs global health surveillance in sub-Saharan Africa'. Code only; no microdata. # Analysis pipeline: data decay and data amnesia R code accompanying *Data decay and data amnesia: what a survey interruption costs global health surveillance in sub-Saharan Africa.* **This repository contains code only.** No microdata, no derived panels, no figures, no manuscript. DHS microdata are not redistributable. See DATA.md for every data source and how to obtain it. ## What the pipeline does It reads DHS recode files for 38 sub-Saharan African countries, computes child mortality, fertility and programme coverage, validates every estimate against the figures DHS publishes, and then asks two questions. How fast does an estimate lose accuracy after the last survey, and how much of an interrupted period can a later survey reconstruct at all. Scope as run: 1,568 recode files, 167 standard DHS rounds, fieldwork 1986 to 2025, plus the survey years of 112 Multiple Indicator Cluster Surveys. ## Pipeline Run in numbered order. Letter suffixes are repair passes and must follow their parent script. **Extraction and validation** | Script | Does | |---|---| | `00_config.R` | Resolves all paths from `DHS_PROJ`. Shared helpers, exclusions, calendar constants. Sourced by every other script. | | `01_extract_mortality_fertility.R` | Birth recodes to under-five, infant and neonatal mortality by synthetic cohort life table. Individual recodes to total fertility. Estimates dated at the midpoint of the period they describe, not at interview. | | `01b_repair_failures.R` | Recovers surveys that fail the primary pass, mainly early phases missing standard design variables. Falls back to a synthetic design using cluster as sampling unit, valid for point estimation only. | | `01c_corrections.R` | Ethiopian calendar offset, applied per survey rather than blanket. | | `02_validate_against_published.R` | Compares every estimate against the DHS Program API. Resolves each survey file to one API year before merging, since `file` is not unique across indicators. | | `03_survey_cadenc …

Visit

github.com

Languages

Amharic

Licenses

MIT

Similaires

Data Decay and Data Amnesia: What a Survey Interruption Costs Global Health Surveillance in Sub-Saharan Africa (analysis code)

Data Decay and Data Amnesia: What a Survey Interruption Costs Global Health Surveillance in Sub-Saharan Africa (analysis code)

R analysis pipeline accompanying the manuscript Data Decay and Data Amnesia: What a Survey Interrupt