Malawi: Incomplete conversion_to_kgs and missing Food Conversion Table

[ligon]

Found in Malawi/_/CONTENTS.org:

1. conversion_to_kgs.json (TODO):
Conversion factors exist for IHS3-IHS5 from wave-specific conversion tables, but there are known issues:

• Item naming doesn't match perfectly between conversion table and food_acquired
• IHS2 has no conversion information at all
• Some items have additional unit detail columns (e.g., '200 GRAMS') that were partially incorporated
• Only consumption and expenditure quantities were converted; other quantities not yet handled

2. Food Conversion Table (TODO):
Not yet created. Note: Malawi data contains caloric data that could be leveraged.

---

Filed from CONTENTS.org audit.

[ligon]

Documented in Malawi/_/CONTENTS.org (both the "conversion_to_kgs.json" and "Food Conversion Table" TODO entries, now tagged with issue #116).

Summary of findings:

• conversion_to_kgs.json has not been generated, despite conversion factor data being available in wave-specific .dta files:
  • 2016-17/Data/Cross_Sectional/com_cf1.dta, com_cf2.dta
  • 2019-20/Data/Cross_Sectional/ihs_foodconversion_factor_2020.dta
  • 2010-11/Documentation/Malawi_IHS3_Food_Item_Conversion_Factors.pdf (methodology)
• Known issues: IHS2 (2004-05) has no conversion data at all; item names don't match perfectly between conversion tables and food_acquired; only consumption/expenditure quantities were converted.
• Food Composition Table (FCT) is also missing, but 2019-20/Data/Cross_Sectional/caloric_conversionfactor.dta provides a head start with caloric data.

[ligon]

Resolved by the post-#231/#223 framework path. Fresh build (NO_CACHE, 2026-06-07): Country('Malawi').food_quantities(units='kgs') resolves 99.9% of rows to u='kg' (970,679 rows), with the residual being non-convertible container units (Basin/Lichero, etc.). Approach C + the #+name: u table replaced the need for a complete conversion_to_kgs. Closing.